02 Cultivation score weights
Goal
Determine the extent to which each of the data elements related to the PG cultivation score (see Ben Porter’s Donor Cultivation Checklist) have impacted giving.
Setup and datafile
library(tidyverse)
library(gridExtra)
library(e1071)
library(wranglR)
library(foreach)
library(splines)
# Functions adapted from previous analysis steps
source('code/functions.R')The data file is generated with this code, adapted from the import data step of 01 Initial exploration.Rmd.
filepath <- 'data/2018-07-26 PG scores for all active prospects.xlsx'
source('code/generate-data.R')Background
The following 12 items were able to be extracted from the database:
- Alum or spouse of alum
- Age
- Chicago home address
- Visited by prospect manager in last 2 years
- 5+ total visits
- Deep engagement (multiple family degrees, parent, season tickets, etc.)
- High-level annual giving ($25K+)
- High-level advisory board participation
- Previous major gift ($250K+)
- Meeting with president
- Open proposal
- Consistent donor (10+ years of giving, including 1+ of last 3)
Each counts as one point toward the cultivation score, which ranges from 0 to 12.
The context for this analysis is to look at how giving is related to the cultivation score. Consider the following two plots (replicated from 01 Initial exploration.Rmd):
There is a nearly linear relationship between PG cultivation score and both campaign giving and an entity’s largest gift or pledge payment. Fitting some sort of linear model will enhance our understanding of the various checklist items by estimating their relative impact.
Data exploration
Indicators
Look at how each of the factors are distributed.
pool %>% select(
ACTIVE_PROPOSALS, AGE, PM_VISIT_LAST_2_YRS, VISITS_5PLUS, AF_25K_GIFT, GAVE_IN_LAST_3_YRS
, MG_250K_PLUS, PRESIDENT_VISIT, TRUSTEE_OR_ADVISORY_BOARD, Alumnus, DEEP_ENGAGEMENT
, CHICAGO_HOME
) %>%
gather('Variable', 'x', 1:12) %>%
group_by(Variable) %>%
summarise(pct.yes = mean(x), yes = sum(x), no = nrow(pool) - sum(x), n = nrow(pool)) %>%
arrange(desc(pct.yes)) %>%
mutate(pct.yes = scales::percent(pct.yes))As expected, alumni status is far and away the leader. I’m a bit surprised that trustee and advisory board is as high as it is. MG and president visit are the least common factors. However, they still have hundreds of observations each so I don’t see a reason to drop either.
Underlying variables
Exploration of the underlying variables for the indicators (that were straightforward to compute). The blue lines indicate the mean, and the purple ones the median.
# Function to produce a summary table, with higher order central moments
summary_moments <- function(x, name = NULL) {
# Requires e1071 for the skewness and kurtosis functions
suppressPackageStartupMessages(
if (!require(e1071)) {
stop('Requires installation of package e1071')
}
)
# If no name passed use the name of x
if (is.null(name)) {name <- quote(x) %>% deparse()}
# Data frame to be returned
data.frame(
name = name
, n = na.omit(x) %>% length()
, min = min(x, na.rm = TRUE)
, median = median(x, na.rm = TRUE)
, mean = mean(x, na.rm = TRUE)
, max = max(x, na.rm = TRUE)
, sd = sd(x, na.rm = TRUE)
, skewness = e1071::skewness(x, na.rm = TRUE)
, kurtosis = e1071::kurtosis(x, na.rm = TRUE)
, NAs = is.na(x) %>% sum()
) %>% return()
}
# Generic function to create histograms
histogrammer <- function(data, x, binwidth = 1, bins = NULL, color = NULL
, trans = 'identity', xbreak = waiver()) {
vec <- paste0('data$', eval(x)) %>% parse(text = .) %>% eval()
data %>%
ggplot(aes_string(x = x, color = color)) +
geom_histogram(binwidth = binwidth, bins = bins, alpha = .5) +
geom_density(aes(y = ..count..), alpha = .5) +
geom_vline(xintercept = mean(vec, na.rm = TRUE), color = 'blue', linetype = 'dashed') +
geom_vline(xintercept = median(vec, na.rm = TRUE), color = 'purple', linetype = 'dotted') +
scale_x_continuous(trans = trans, breaks = xbreak)
}pool %>% filter(!is.na(NUMERIC_AGE)) %>%
histogrammer(x = 'NUMERIC_AGE', xbreak = seq(0, 200, by = 10)) +
labs(title = 'Age')
summary_moments(pool$NUMERIC_AGE, 'Age')The age distribution is moderately right-skewed but looks fine. For a quick fix the NAs could be imputed as the group mean.
pool %>%
histogrammer(x = 'VISIT_COUNT') +
labs(title = 'Visit count')
summary_moments(pool$VISIT_COUNT, 'Visits')It’d be sensible to transform the x axis or otherwise get rid of those outliers.
pool %>%
histogrammer(x = 'VISIT_COUNT', trans = 'sqrt', binwidth = NULL, bins = 200,
xbreak = c(seq(0, 10, by = 2), seq(10, 100, by = 10), seq(100, 200, by = 20))) +
labs(title = 'Sqrt visit, log10 count') +
scale_y_continuous(trans = 'log10plus1', breaks = 10^(0:20))
Nearly a linear decrease in visit count on a log/sqrt scale - though I’m not sure how to interpret this.
pool %>%
histogrammer(x = 'YEARS_OF_GIVING') +
labs(title = 'Years of giving')
summary_moments(pool$YEARS_OF_GIVING, 'Years of giving')This looks fine. There are more loyal donors than I would’ve thought.
pool %>%
histogrammer(x = 'YEARS_OF_GIVING_LAST_3') +
labs(title = 'Years of giving last 3')
summary_moments(pool$YEARS_OF_GIVING_LAST_3, 'Years of giving out of last 3')Mostly nondonors, unsurprisingly.
pool %>%
histogrammer(x = 'MG_250K_COUNT') +
labs(title = 'Count of $250K+ major gifts') +
scale_y_continuous(trans = 'log10plus1', breaks = 10^(0:5), labels = 10^(0:5))
summary_moments(pool$MG_250K_COUNT, '$250K+ major gifts')pool %>% group_by(MG_250K_COUNT) %>% mutate(n = 1) %>%
summarise(
total = sum(n)
, proportion = {sum(n) / nrow(pool)} %>% scales::percent()
)Extremely few people make a single $250K+ gift, much less multiple ones.
pool %>%
histogrammer(x = 'SEASON_TICKET_YEARS', xbreak = seq(0, 20, by = 2)) +
labs(title = 'Years holding season tickets') +
scale_y_continuous(trans = 'log10plus1', breaks = 10^(0:20))
summary_moments(pool$SEASON_TICKET_YEARS, 'Years holding season tickets')That jump at 10 years is odd. Perhaps season tickets have only been consistently tracked for about 10 years?
Underlying variable scatterplots
pool %>% filter(!is.na(NUMERIC_AGE)) %>%
scatterplotter(x = 'NUMERIC_AGE', y = 'CAMPAIGN_NEWGIFT_CMIT_CREDIT', color = 'MG_PR_MODEL_DESC'
, ytrans = 'log10plus1', ylabels = scales::dollar) +
geom_vline(aes(xintercept = mean(NUMERIC_AGE)), color = 'blue', linetype = 'dashed') +
labs(title = 'Age versus campaign giving')
pool %>% filter(!is.na(NUMERIC_AGE)) %>%
scatterplotter(x = 'NUMERIC_AGE', y = 'LARGEST_GIFT_OR_PAYMENT', color = 'MG_PR_MODEL_DESC'
, ytrans = 'log10plus1', ylabels = scales::dollar) +
geom_vline(aes(xintercept = mean(NUMERIC_AGE)), color = 'blue', linetype = 'dashed') +
labs(title = 'Age versus largest gift')
As usual, age is positively associated with giving. The outlier 113-year-old should probably be removed.
max_age <- 110
pool %>% filter(NUMERIC_AGE <= max_age) %>%
scatterplotter(x = 'NUMERIC_AGE', y = 'LARGEST_GIFT_OR_PAYMENT'
, color = 'MG_PR_MODEL_DESC', ytrans = 'log10plus1', ylabels = scales::dollar) +
geom_vline(aes(xintercept = mean(NUMERIC_AGE)), color = 'blue', linetype = 'dashed') +
labs(title = bquote('Age versus largest gift' ~ (age <= .(max_age)) ))
For visits, based on the above exploration visit count needs a transformation.
pool %>%
scatterplotter(x = 'VISIT_COUNT', y = 'CAMPAIGN_NEWGIFT_CMIT_CREDIT', color = 'MG_PR_MODEL_DESC'
, ytrans = 'log10plus1', ylabels = scales::dollar) +
geom_vline(aes(xintercept = mean(VISIT_COUNT)), color = 'blue', linetype = 'dashed') +
labs(title = 'Log-sqrt visit count versus campaign giving') +
scale_x_sqrt()
pool %>%
scatterplotter(x = 'VISIT_COUNT', y = 'LARGEST_GIFT_OR_PAYMENT', color = 'MG_PR_MODEL_DESC'
, ytrans = 'log10plus1', ylabels = scales::dollar) +
geom_vline(aes(xintercept = mean(VISIT_COUNT)), color = 'blue', linetype = 'dashed') +
labs(title = 'Log-sqrt visit count versus largest gift') +
scale_x_sqrt()
pool %>%
scatterplotter(x = 'VISIT_COUNT', y = 'CAMPAIGN_NEWGIFT_CMIT_CREDIT', color = 'MG_PR_MODEL_DESC'
, ytrans = 'log10plus1', ylabels = scales::dollar) +
geom_vline(aes(xintercept = mean(VISIT_COUNT)), color = 'blue', linetype = 'dashed') +
labs(title = 'Log-log visit count versus campaign giving') +
scale_x_continuous(trans = 'log10plus1', breaks = c(0, 1, 10, 50, 100, 150, 200))
pool %>%
scatterplotter(x = 'VISIT_COUNT', y = 'LARGEST_GIFT_OR_PAYMENT', color = 'MG_PR_MODEL_DESC'
, ytrans = 'log10plus1', ylabels = scales::dollar) +
geom_vline(aes(xintercept = mean(VISIT_COUNT)), color = 'blue', linetype = 'dashed') +
labs(title = 'Log-log visit count versus largest gift') +
scale_x_continuous(trans = 'log10plus1', breaks = c(0, 1, 10, 50, 100, 150, 200))
The log-log plots look quite good. Visits \(\geq\) 100 are outliers, but at first glance don’t appear influential on the log-log scale (easy to test with e.g. Cook’s D).
pool %>%
scatterplotter(x = 'YEARS_OF_GIVING', y = 'CAMPAIGN_NEWGIFT_CMIT_CREDIT', color = 'MG_PR_MODEL_DESC'
, ytrans = 'log10plus1', ylabels = scales::dollar) +
geom_vline(aes(xintercept = mean(YEARS_OF_GIVING)), color = 'blue', linetype = 'dashed') +
labs(title = 'Years of giving versus campaign giving') +
scale_x_sqrt()
pool %>%
scatterplotter(x = 'YEARS_OF_GIVING', y = 'LARGEST_GIFT_OR_PAYMENT', color = 'MG_PR_MODEL_DESC'
, ytrans = 'log10plus1', ylabels = scales::dollar) +
geom_vline(aes(xintercept = mean(YEARS_OF_GIVING)), color = 'blue', linetype = 'dashed') +
labs(title = 'Years of giving versus largest gift') +
scale_x_sqrt()
The square root transformation does well for campaign giving, but not as well for largest gift or payment.
# Box-Cox test for transformations
boxcox_lambdas <- seq(-1, 1, by = .01)
boxcox_lm <- lm(I(LARGEST_GIFT_OR_PAYMENT + 1) ~ YEARS_OF_GIVING, data = pool) %>%
MASS::boxcox(lambda = boxcox_lambdas, plotit = FALSE)
maxLL <- boxcox_lm$x[which(boxcox_lm$y == max(boxcox_lm$y))]
# Plot results
boxcox_lm %>%
unlist() %>%
matrix(nrow = length(boxcox_lambdas)) %>%
data.frame() %>%
select(x = X1, y = X2) %>%
ggplot(aes(x = x, y = y)) +
geom_line() +
geom_vline(aes(xintercept = x[which(y == max(y))]), color = 'blue', linetype = 'dashed') +
labs(title = 'Box-Cox analysis', x = expression(lambda), y = 'log Likelihood')
\(\lambda =\) 0.03 is pretty close to a log transformation.
# Best Box-Cox transformation, adding 1 so the response variable is strictly positive
boxcoxbest_trans <- function(x) {
scales::trans_new(
'boxcoxbest'
, transform = function(x) {(x + 1)^maxLL}
, inverse = function(x) {(x + 1)^(1/maxLL)}
)
}grid.arrange(
# Plot Box-Cox results
pool %>%
scatterplotter(x = 'YEARS_OF_GIVING', y = 'LARGEST_GIFT_OR_PAYMENT', color = 'MG_PR_MODEL_DESC'
, ytrans = 'log10plus1', ylabels = scales::dollar) +
geom_vline(aes(xintercept = mean(YEARS_OF_GIVING)), color = 'blue', linetype = 'dashed') +
labs(title = 'Years of giving versus campaign giving, Box-Cox transformation', y = 'Largest gift') +
scale_x_continuous(trans = 'boxcoxbest', breaks = c(0, 10))
# Plot log10 results
, pool %>%
scatterplotter(x = 'YEARS_OF_GIVING', y = 'LARGEST_GIFT_OR_PAYMENT', color = 'MG_PR_MODEL_DESC'
, ytrans = 'log10plus1', ylabels = scales::dollar) +
geom_vline(aes(xintercept = mean(YEARS_OF_GIVING)), color = 'blue', linetype = 'dashed') +
labs(title = 'Years of giving versus campaign giving, log transformation', y = 'Largest gift') +
scale_x_continuous(trans = 'log10plus1')
)
After all that neither look linear, though they do look nearly linear around the mean.
pool %>% mutate(last_3_yrs = factor(YEARS_OF_GIVING_LAST_3)) %>%
ggplot(aes(x = last_3_yrs, y = CAMPAIGN_NEWGIFT_CMIT_CREDIT, color = MG_PR_MODEL_DESC)) +
geom_boxplot() +
facet_grid(. ~ MG_PR_MODEL_DESC) +
scale_y_continuous(trans = 'log10plus1', labels = scales::dollar, breaks = 10^(0:20)) +
labs(title = 'Years of giving of last 3 versus campaign giving')
pool %>% mutate(last_3_yrs = factor(YEARS_OF_GIVING_LAST_3)) %>%
ggplot(aes(x = last_3_yrs, y = LARGEST_GIFT_OR_PAYMENT, color = MG_PR_MODEL_DESC)) +
geom_boxplot() +
facet_grid(. ~ MG_PR_MODEL_DESC) +
scale_y_continuous(trans = 'log10plus1', labels = scales::dollar, breaks = 10^(0:20)) +
labs(title = 'Years of giving of last 3 versus largest gift')
I see a main effect for campaign giving.
pool %>%
scatterplotter(x = 'MG_250K_COUNT', y = 'CAMPAIGN_NEWGIFT_CMIT_CREDIT', color = 'MG_PR_MODEL_DESC'
, ytrans = 'log10plus1', ylabels = scales::dollar) +
geom_vline(aes(xintercept = mean(MG_250K_COUNT)), color = 'blue', linetype = 'dashed') +
labs(title = 'Count of $250K+ gifts versus campaign giving') +
scale_x_sqrt()
pool %>%
scatterplotter(x = 'MG_250K_COUNT', y = 'LARGEST_GIFT_OR_PAYMENT', color = 'MG_PR_MODEL_DESC'
, ytrans = 'log10plus1', ylabels = scales::dollar) +
geom_vline(aes(xintercept = mean(MG_250K_COUNT)), color = 'blue', linetype = 'dashed') +
labs(title = 'Count of $250K+ gifts versus largest gift') +
scale_x_sqrt()
As seen above, pretty much all of the observations are outliers. Might make sense to leave this one discretized.
pool %>%
scatterplotter(x = 'SEASON_TICKET_YEARS', y = 'CAMPAIGN_NEWGIFT_CMIT_CREDIT', color = 'MG_PR_MODEL_DESC'
, ytrans = 'log10plus1', ylabels = scales::dollar) +
geom_vline(aes(xintercept = mean(SEASON_TICKET_YEARS)), color = 'blue', linetype = 'dashed') +
labs(title = 'Season ticket years versus campaign giving') +
scale_x_continuous(breaks = seq(0, 100, by = 2))
pool %>%
scatterplotter(x = 'SEASON_TICKET_YEARS', y = 'LARGEST_GIFT_OR_PAYMENT', color = 'MG_PR_MODEL_DESC'
, ytrans = 'log10plus1', ylabels = scales::dollar) +
geom_vline(aes(xintercept = mean(SEASON_TICKET_YEARS)), color = 'blue', linetype = 'dashed') +
labs(title = 'Season ticket years versus largest gift') +
scale_x_continuous(breaks = seq(0, 100, by = 2))
This looks fine.
Data exploration takeaways
- Missing age values can be imputed as mean age
- The 113-year-old is an outlier
- Visit count could use log transform
- Years of giving could use sqrt transform
- $250K+ gifts is a binary indicator
Modeling ground rules
I’ll outline some basic methodology in advance of modeling to try to account for researcher degrees of freedom.
As seen above, both campaign giving and largest lifetime gift have a strong linear relationship with the cultivation score, formulated as a 0 to 12 point scale. Currently, all criteria count as a single point (equally weighted) but I suspect they impact actual giving behavior differently. I propose four models:
- Linear regression of campaign giving on cultivation score variables
- Linear regression of campaign on all available explanatory variables, essentially looking at the cultivation score variables controlling for the preexisting modeled scores
- Linear regression of largest cash transaction on cultivation score variables
- Linear regression of largest cash transaction on all available explanatory variables, again to control for the preexisting modeled scores
If a given explanatory variable has the same coefficient sign and similar nonzero magnitudes in all models, I’ll take this as evidence of an association with giving behavior.
After withholding a random 20% of the data as a test set, I’ll use ten-fold cross-validation to determine which variables should be included in each model. Model performance will be determined based on out-of-sample mean squared error, defined as usual:
\[ {\text{MSE}} = \frac{1}{n} \sum_{i=1}^{n} \left( Y_i - \hat{Y}_i \right)^2 \]
The 20% test set will be used to confirm the cross-validation results, and then final models will be constructed on the entire dataset for comparison.
Data cleanup
First, perform some quick data clean-up.
mdat <- pool %>% mutate(
# Impute missing ages as mean age
NUMERIC_AGE = case_when(
!is.na(NUMERIC_AGE) ~ NUMERIC_AGE
, TRUE ~ mean(NUMERIC_AGE, na.rm = TRUE)
)
# Impute missing affinity scores as mean affinity score
, AFFINITY_SCORE = case_when(
!is.na(AFFINITY_SCORE) ~ AFFINITY_SCORE
, TRUE ~ mean(AFFINITY_SCORE, na.rm = TRUE)
)
# Create null factor levels for the MG_ID and MG_PR models
, MG_ID_MODEL_DESC = fct_explicit_na(MG_ID_MODEL_DESC, 'Unscored') %>% fct_relevel('Unscored')
, MG_PR_MODEL_DESC = fct_explicit_na(MG_PR_MODEL_DESC, 'Unscored') %>% fct_relevel('Unscored')
# Create row numbers
, rownum = 1:nrow(pool)
) %>% select(
# Drop unhelpful fields
-ID_NUMBER, -PROSPECT_ID, -PROSPECT_NAME, -NU_DEG, -NU_DEG_SPOUSE, -POTENTIAL_INTEREST_AREAS
, -PREF_NAME_SORT, -MG_ID_MODEL_YEAR, -MG_ID_MODEL_SCORE, -MG_PR_MODEL_YEAR, -MG_PR_MODEL_SCORE
)I’ll use my wranglR::KFoldXVal function to create the cross-validation groups.
k <- 11
# Create k groups: first is 20% of the data (prop = .2) and the others are equally sized
xval_inds <- pool %>% wranglR::KFoldXVal(k = k, prop = .2, seed = 12644)
# Remove outliers
outliers <- which(mdat$NUMERIC_AGE > max_age)
for (i in 1:k) {
rm_idx <- which(xval_inds[[i]] %in% outliers)
if (length(rm_idx) > 0) {xval_inds[[i]] <- xval_inds[[i]][-rm_idx]}
}
# Create groups
oos_inds <- xval_inds[[1]]
xval_inds <- xval_inds[2:k]
# Results
c(list(oos_inds), xval_inds) %>% summary Length Class Mode
[1,] 5215 -none- numeric
[2,] 2086 -none- numeric
[3,] 2086 -none- numeric
[4,] 2086 -none- numeric
[5,] 2086 -none- numeric
[6,] 2086 -none- numeric
[7,] 2086 -none- numeric
[8,] 2086 -none- numeric
[9,] 2086 -none- numeric
[10,] 2086 -none- numeric
[11,] 2090 -none- numericGroup 1 is the out-of-sample validation set, and the other 10 will be used for cross-validation.
Finally, I’ll create a quick function to compute MSE.
calc_mse <- function(y, yhat) {
mean(
(y - yhat)^2, na.rm = TRUE
)
}Visualization functions
# Create coefficients data frame
create_coefs <- function(model_list) {
foreach(i = 1:length(model_list), .combine = 'rbind') %do% {
tmp <- summary(model_list[[i]])$coefficients
data.frame(tmp) %>%
mutate(
variable = rownames(tmp)
, model = i
) %>% select(
model
, variable
, beta.hat = Estimate
, SE = Std..Error
, t.val = t.value
, Pr.t = Pr...t..
) %>% return()
} %>% return()
}
# Plot R-squared
plot_r2 <- function(model_list, type = 'r.squared') {
parser <- function(x) {
tmpsum <- summary(x)
paste0('tmpsum$', type) %>% parse(text = .) %>% eval() %>% return()
}
model_list %>%
lapply(., function(x) parser(x)) %>% unlist() %>% data.frame(r.squared = .) %>%
ggplot(aes(x = r.squared)) +
geom_density() +
geom_vline(aes(xintercept = mean(r.squared)), color = 'blue', linetype = 'dashed', alpha = .5) +
geom_rug(color = 'blue') +
labs(title = bquote('Density plot of' ~ r^2 ~ 'results, mean' ~
.(lapply(model_list, function(x) {summary(x)$r.squared}) %>% unlist() %>% mean() %>% round(3))
)
, x = bquote(r^2)
)
}
# Plot cross-validated coefficients
plot_coefs <- function(model_list, conf_interval = 1 - p.sig) {
crit_val <- qnorm({1 - conf_interval} / 2) %>% abs()
coefs <- create_coefs(model_list)
coefs %>% full_join(
coefs %>% group_by(variable) %>%
summarise(group.mean = mean(beta.hat), group.sd = sd(beta.hat))
, by = c('variable', 'variable')
) %>%
ggplot(aes(x = variable, y = beta.hat, color = factor(model))) +
geom_segment(
aes(
xend = variable
, y = group.mean - 2 * crit_val * group.sd
, yend = group.mean + 2 * crit_val * group.sd
), color = 'gray', alpha = .25, size = 2) +
geom_point() +
geom_hline(yintercept = 0, alpha = .5) +
theme(axis.text.x = element_text(angle = 90, hjust = 1, vjust = .5), axis.title.y = element_text(angle = 0, vjust = .5)) +
labs(
title = 'Coefficient estimates per cross-validation model'
, y = bquote(hat(beta))
, color = 'cross-validation sample'
)
}
# Table of coefficient +/- counts
coef_pm_table <- function(model_list, pval) {
create_coefs(model_list) %>%
group_by(variable) %>%
summarise(
`+` = sum(sign(beta.hat) == 1 & Pr.t < pval)
, `0` = sum(Pr.t >= pval)
, `-` = sum(sign(beta.hat) < 0 & Pr.t < pval)
)
}
# Compute predictions
calc_preds <- function(model_list, xval, yname) {
yhats <- list()
for (i in 1:length(model_list)) {
yhats[[i]] <- data.frame(
model = i
, row = xval[[i]]
, preds = model_list[[i]] %>% predict(newdata = mdat[xval[[i]], ])
, truth = mdat[xval[[i]], yname] %>% unlist() %>% log10plus1()
)
}
return(yhats)
}
calc_outsample_mse <- function(model_list, xval, yname) {
calc_preds(model_list, xval, yname) %>%
lapply(function(x) calc_mse(y = x$truth, yhat = x$preds)) %>%
unlist()
}
# Plot MSEs by insample/outsample
plot_mses <- function(model_list, xval, truth) {
mses <- data.frame(
insample = model_list %>%
lapply(function(x) calc_mse(y = model.frame(x)[, 1], yhat = predict(x))) %>%
unlist()
, outsample = calc_outsample_mse(model_list, xval, truth)
) %>% gather('type', 'MSE', 1:2)
mses %>%
ggplot(aes(x = MSE, color = type)) +
geom_density() +
geom_vline(
xintercept = mses %>% filter(type == 'insample') %>% select(MSE) %>% unlist %>% mean()
, color = 'red', linetype = 'dashed', alpha = .5
) +
geom_vline(
xintercept = mses %>% filter(type == 'outsample') %>% select(MSE) %>% unlist %>% mean()
, color = 'darkcyan', linetype = 'dashed', alpha = .5
) +
geom_rug() +
labs(
title = bquote('MSE across samples, means =' ~
.(mses %>% group_by(type) %>% summarise(mean = mean(MSE)) %>%
select(mean) %>% unlist() %>% round(3) %>% paste(collapse = ', ')
)
)
)
}
# Merges predicted results into one large data frame each for insample and outsample
calc_resids <- function(model_list, xval, yname) {
insample <- foreach(i = 1:length(model_list), .combine = 'rbind') %do% {
data.frame(
model = i
, preds = model_list[[i]] %>% predict()
, truth = model.frame(model_list[[i]])[, 1]
) %>% mutate(
residuals = truth - preds
)
}
preds <- calc_preds(model_list, xval, yname)
outsample <- foreach(i = 1:length(model_list), .combine = 'rbind') %do% {
preds[[i]]
} %>% mutate(
residuals = truth - preds
)
return(list(insample = insample, outsample = outsample))
}
# Plot standardized residuals; returns a list of ggplot objects $insample and $outsample
plot_resids <- function(model_list, xval, yname, filter = 'TRUE') {
resids <- calc_resids(model_list, xval, yname)
# Plot residuals vs fitted for in-sample data
insample <- resids$insample %>% filter_(filter) %>%
ggplot(aes(x = preds, y = residuals, color = factor(model))) +
geom_point(alpha = .01) +
geom_smooth(se = FALSE) +
labs(title = 'In-sample residuals versus fitted', color = 'cross-validation sample')
# Plot residuals vs fitted for out-of-sample data
outsample <- resids$outsample %>% filter_(filter) %>%
ggplot(aes(x = preds, y = residuals, color = factor(model))) +
geom_point(alpha = .1) +
geom_smooth(se = FALSE) +
labs(title = 'Out-of-sample residuals versus fitted', color = 'cross-validation sample')
return(list(insample = insample, outsample = outsample))
}
# Plot normal Q-Q visualization for residuals
plot_qq <- function(model_list, xval, yname, filter = 'TRUE') {
resids <- calc_resids(model_list, xval, yname)
# In-sample Q-Q plot with standardized residuals
insample <- resids$insample %>% mutate(st.resid = residuals/sd(residuals)) %>% filter_(filter) %>%
ggplot(aes(sample = st.resid, color = factor(model))) +
geom_qq(alpha = .05) +
geom_qq_line() +
labs(title = 'In-sample Q-Q plot with standardized residuals'
, color = 'cross-validation sample')
# Out-of-sample Q-Q plot
outsample <- resids$outsample %>% mutate(st.resid = residuals/sd(residuals)) %>% filter_(filter) %>%
ggplot(aes(sample = st.resid, color = factor(model))) +
geom_qq(alpha = .05) +
geom_qq_line() +
labs(title = 'Out-of-sample Q-Q plot with standardized residuals'
, color = 'cross-validation sample')
return(list(insample = insample, outsample = outsample))
}
# Compute partial residuals
calc_partial_resids <- function(model, inds, xname, df = NULL) {
# Quick hack to get around update() environment, not really recommended
scoped_xname <<- xname
scoped_spline_df <<- df
# Partial residuals data frame
data.frame(
x = mdat %>% filter(rownum %in% unlist(inds)) %>% select_(xname) %>% unlist()
, y = model.frame(model)[, 1]
, y.hat = if (!is.null(df)) {
update(model, formula = . ~ ns(parse(text = scoped_xname) %>% eval(), df = scoped_spline_df)) %>% fitted()
} else {
update(model, formula = . ~ parse(text = scoped_xname) %>% eval()) %>% fitted()
}
) %>% mutate(
# Compute partial residuals
y.partial.resid = if (!is.null(df)) {
{model %>% residuals(type = 'partial')}[, paste0('ns(', parse(text = xname), ', df = ', df, ')')]
} else {
{model %>% residuals(type = 'partial')}[, xname]
}
# Regression on partial residuals
, y.hat.partial = if (!is.null(df)) {
lm(y.partial.resid ~ ns(x, df)) %>% fitted()
} else {
lm(y.partial.resid ~ x) %>% fitted()
}
) %>% return()
}Campaign linear models
Cultivation score predictors
Regress campaign giving against each of the cultivation score predictors.
# List to store campaign linear models
clms <- list()
for(i in 1:length(xval_inds)) {
# Create linear model excluding the holdout and out-of-sample indices
clms[[i]] <- mdat %>%
filter(rownum %in% unlist(xval_inds)[unlist(xval_inds) %nin% xval_inds[[i]]]) %>%
lm(
log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~
ACTIVE_PROPOSALS +
AGE +
PM_VISIT_LAST_2_YRS +
VISITS_5PLUS +
AF_25K_GIFT +
GAVE_IN_LAST_3_YRS +
MG_250K_PLUS +
PRESIDENT_VISIT +
TRUSTEE_OR_ADVISORY_BOARD +
Alumnus +
DEEP_ENGAGEMENT +
CHICAGO_HOME
, data = .
)
}The full (and hard to read) results for each model are in the appendix.
We can extract a few parameters of interest.
plot_r2(clms) +
geom_text(y = seq(10, 100, length.out = k - 1), label = 1:(k - 1), color = 'blue') +
xlim(c(.45, .48))
The average \(r^2 =\) 0.465 is quite a good result.
p.sig <- 1E-2
plot_coefs(clms)
coef_pm_table(clms, p.sig)The coefficients are extremely tightly clustered within each cross-validation set. Interestingly, age and alumni status both have negative coefficients. All are significant at \(p =\) 0.01.
Here are the actual prediction MSEs:
plot_mses(clms, xval_inds, 'CAMPAIGN_NEWGIFT_CMIT_CREDIT')
As usual, in-sample performance is moderately optimistic.
plot_resids(clms, xval_inds, 'CAMPAIGN_NEWGIFT_CMIT_CREDIT')$insample +
scale_y_continuous(breaks = seq(-10, 10, by = 2))
plot_resids(clms, xval_inds, 'CAMPAIGN_NEWGIFT_CMIT_CREDIT')$outsample +
scale_y_continuous(breaks = seq(-10, 10, by = 2))
At first glance, the linear trend evident in these results is concerning. Upon second glance, while still sobering it really just reinforces what we’d observed in the first of the two plots in the Background section: campaign giving is decidedly nonlinear as cultivation scores approach their high/low limits. As seen above, the underlying factors should be transformed, and possibly modeled nonlinearly (splines).
plot_qq(clms, xval_inds, 'CAMPAIGN_NEWGIFT_CMIT_CREDIT')$insample +
scale_y_continuous(breaks = seq(-10, 10, by = 2))
plot_qq(clms, xval_inds, 'CAMPAIGN_NEWGIFT_CMIT_CREDIT')$outsample +
scale_y_continuous(breaks = seq(-10, 10, by = 2))
The quantile-quantile plot further illustrates the issue. The drift above the reference line to the left and below it to the right suggests less density than expected in the tails, which follows given that the range is bound – it’s not possible to give less than a nondonor or (for practical purposes) more than a 9-figure donor.
The story is completely different when looking only at those who actually gave:
plot_qq(clms, xval_inds, 'CAMPAIGN_NEWGIFT_CMIT_CREDIT', filter = 'truth > 0')$insample +
scale_y_continuous(breaks = seq(-10, 10, by = 2))
plot_qq(clms, xval_inds, 'CAMPAIGN_NEWGIFT_CMIT_CREDIT', filter = 'truth > 0')$outsample +
scale_y_continuous(breaks = seq(-10, 10, by = 2))
This is quite a bit nicer, suggesting only slight skewness in the tails.
Takeaway. When modeling giving amounts in the future, consider fitting a conditional model. Something along the lines of:
\[ E(\text{giving amount} ~ | ~ \text{donor status} = 1) \]
All predictors 1
The second model compares the variables underlying each PG score indicator, plus supplemental predictors. The first model incudes everything and I’ll prune from there using MSE. Initial spline df are fairly arbitrary.
# List to store campaign linear models
clmaps <- list()
for(i in 1:length(xval_inds)) {
# Create linear model excluding the holdout and out-of-sample indices
clmaps[[i]] <- mdat %>%
filter(rownum %in% unlist(xval_inds)[unlist(xval_inds) %nin% xval_inds[[i]]]) %>%
lm(
log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~
ACTIVE_PROPOSALS +
ns(NUMERIC_AGE, df = 5) + # Underlying variable to AGE indicator
PM_VISIT_LAST_2_YRS +
log10plus1(VISIT_COUNT) + # Underlying VISITS_5PLUS indicator
AF_25K_GIFT +
sqrt(YEARS_OF_GIVING) + # Underlying GAVE_IN_LAST_3_YRS
ns(YEARS_OF_GIVING_LAST_3, df = 2) + # Underlying GAVE_IN_LAST_3_YRS
MG_250K_PLUS + # Decided to leave as factor
PRESIDENT_VISIT +
TRUSTEE_OR_ADVISORY_BOARD +
Alumnus +
DOUBLE_ALUM + # Deep Engagement component
EVER_PARENT + # Deep Engagement component
SEASON_TICKET_YEARS + # Deep Engagement component
CHICAGO_HOME +
QUAL_LEVEL +
AFFINITY_SCORE +
MG_PR_MODEL_DESC
, data = .
)
}Full results are in the appendix.
plot_r2(clmaps) +
geom_text(y = seq(10, 150, length.out = k - 1), label = 1:(k-1), color = 'blue') +
xlim(c(.72, .735))
This is a much higher \(r^2\) than seen above, but the model also includes many more predictors. Consider the MSE.
plot_mses(clmaps, xval_inds, 'CAMPAIGN_NEWGIFT_CMIT_CREDIT')
Well, that’s pretty conclusive – this is also much lower than that seen previously. This implies that on average the predicted giving amount is less than a factor of 10 off. Which predictors contribute to the performance?
plot_coefs(clmaps)
coef_pm_table(clmaps, p.sig)After accounting for the other variables, things that don’t seem to matter include active proposals, Chicago home address, two of the deep engagement indicators, president visits, qualification level (inconsistent), and committee participation. Note that the “Future Prospect” factor level only appears 9 times – one of the cross-validation samples must not have had anyone rated at that level. Additionally, some of these are likely already included in the affinity score.
plot_resids(clmaps, xval_inds, 'CAMPAIGN_NEWGIFT_CMIT_CREDIT')$outsample +
scale_y_continuous(breaks = seq(-10, 10, by = 2))
plot_qq(clmaps, xval_inds, 'CAMPAIGN_NEWGIFT_CMIT_CREDIT')$outsample 
This already looks a lot nicer than the Cultivation score predictors model! Now let’s try again, dropping the less-interesting predictors.
All predictors 2
Full results are in the appendix.
# List to store campaign linear models
clmaps2 <- list()
for(i in 1:length(xval_inds)) {
# Create linear model excluding the holdout and out-of-sample indices
clmaps2[[i]] <- mdat %>%
filter(rownum %in% unlist(xval_inds)[unlist(xval_inds) %nin% xval_inds[[i]]]) %>%
lm(
log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~
ns(NUMERIC_AGE, df = 5) + # Underlying variable to AGE indicator
PM_VISIT_LAST_2_YRS +
log10plus1(VISIT_COUNT) + # Underlying VISITS_5PLUS indicator
AF_25K_GIFT +
sqrt(YEARS_OF_GIVING) + # Underlying GAVE_IN_LAST_3_YRS
ns(YEARS_OF_GIVING_LAST_3, df = 2) + # Underlying GAVE_IN_LAST_3_YRS
MG_250K_PLUS + # Decided to leave as factor
Alumnus +
SEASON_TICKET_YEARS + # Deep Engagement component
AFFINITY_SCORE +
MG_PR_MODEL_DESC
, data = .
)
}plot_mses(clmaps2, xval_inds, 'CAMPAIGN_NEWGIFT_CMIT_CREDIT')
That’s a marginal increase in outsample MSE (0.9211 to 0.9316, difference of 1.14%) with many fewer predictors.
plot_coefs(clmaps2)
coef_pm_table(clmaps2, p.sig)The spline degrees of freedom for numeric age could be tweaked.
All predictors splines test
splines_max <- 10
clmaps3 <- list()
for (s in 1:splines_max) {
clmaps3[[s]] <- list()
for(i in 1:length(xval_inds)) {
# Create linear model excluding the holdout and out-of-sample indices
clmaps3[[s]][[i]] <- mdat %>%
filter(rownum %in% unlist(xval_inds)[unlist(xval_inds) %nin% xval_inds[[i]]]) %>%
lm(
log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~
ns(NUMERIC_AGE, df = s) + # Underlying variable to AGE indicator
PM_VISIT_LAST_2_YRS +
log10plus1(VISIT_COUNT) + # Underlying VISITS_5PLUS indicator
AF_25K_GIFT +
sqrt(YEARS_OF_GIVING) + # Underlying GAVE_IN_LAST_3_YRS
ns(YEARS_OF_GIVING_LAST_3, df = 2) + # Underlying GAVE_IN_LAST_3_YRS
MG_250K_PLUS + # Decided to leave as factor
Alumnus +
SEASON_TICKET_YEARS + # Deep Engagement component
AFFINITY_SCORE +
MG_PR_MODEL_DESC
, data = .
)
}
}Try different spline degrees of freedom for NUMERIC_AGE. Full results are in the appendix.
Consider the distribution of MSEs for each model.
# Calculate MSEs for each model
spline_mse <- foreach(s = 1:splines_max, .combine = rbind) %do% {
mses <- calc_outsample_mse(clmaps3[[s]], xval_inds, 'CAMPAIGN_NEWGIFT_CMIT_CREDIT')
data.frame(spline.df = s, xv_group = factor(1:(k-1)), mses)
}# Plot results
spline_mse %>%
ggplot(aes(x = spline.df, y = mses)) +
geom_point(aes(color = xv_group)) +
geom_smooth() +
scale_x_continuous(breaks = 1:splines_max, minor_breaks = NULL) +
labs(x = 'spline df', y = 'MSE', color = 'cross-validation sample')
For practical purposes there’s not much difference between the different choices. It looks like 4 or 5 is where the mean MSE levels out, so I’ll stick with my initial choice.
Comparison
Create the two final models and check them on out-of-sample data.
# Predict campaign giving, PG score indicators
clm_final <- clms[[1]] %>% update(data = mdat %>% filter(rownum %in% unlist(xval_inds)))
# Predict campaign giving, underlying factors
clmap_final <- clmaps2[[1]] %>% update(data = mdat %>% filter(rownum %in% unlist(xval_inds)))Full results in the appendix
plot_resids(list(clmap_final, clm_final), list(oos_inds, oos_inds), 'CAMPAIGN_NEWGIFT_CMIT_CREDIT')$outsample +
scale_y_continuous(breaks = seq(-10, 10, by = 2)) +
scale_color_discrete(labels = c('All predictors', 'PG indicators')) +
labs(color = 'Model')
The “All predictors” residuals look better than the “PG indicators”" residuals.
plot_qq(list(clmap_final, clm_final), list(oos_inds, oos_inds), 'CAMPAIGN_NEWGIFT_CMIT_CREDIT')$outsample +
scale_y_continuous(breaks = seq(-10, 10, by = 2)) +
scale_color_discrete(labels = c('All predictors', 'PG indicators')) +
labs(color = 'Model')
Note the reference lines – “All predictors” is much closer to a normal distribution, albeit with some positive skewness.
c_mses_final <- rbind(
calc_preds(list(clm_final), list(oos_inds), 'CAMPAIGN_NEWGIFT_CMIT_CREDIT')[[1]] %>%
mutate(model = 'PG indicators')
, calc_preds(list(clmap_final), list(oos_inds), 'CAMPAIGN_NEWGIFT_CMIT_CREDIT')[[1]] %>%
mutate(model = 'All predictors')
) %>% mutate(
model = factor(model)
, sq.error = (truth - preds)^2
)
c_mses_final <- c_mses_final %>% left_join(
c_mses_final %>% group_by(model) %>% summarise(mse = mean(sq.error))
, by = c('model', 'model')
)
c_mses_final %>%
ggplot(aes(x = sq.error, color = model)) +
geom_density() +
geom_vline(aes(xintercept = mean(sq.error)), linetype = 'dotted', alpha = .5) +
geom_vline(aes(xintercept = mse, color = model), linetype = 'dashed') +
xlim(c(0, 5)) +
labs(
x = 'squared error'
, title = bquote('Squared error across models, means = ' ~
.(c_mses_final %>% group_by(model) %>% summarise(mse = mean(mse)) %>% select(mse) %>%
unlist() %>% round(3) %>% paste(collapse = ', ')))
)
We never want to see bimodal squared errors – it looks like “All predictors” is a much better regression model. These are all very close to the results seen in the corresponding sections above, which is reassuring.
Finally, let’s look at the impact of age in the splines model. I’ll try plotting against both the raw \(\textbf{y}\) and the partial residuals, which are obtained by removing the effects of all the predictors besides the \(\textbf{x}_j\) we’re interested in, e.g.
\[ \boldsymbol{\hat{\epsilon}} = \textbf{y} - \hat{\textbf{y}} = \textbf{y} - \sum_{i} {\textbf{x}_i \hat{\boldsymbol{\beta}}_i} \] \[ \boldsymbol{\hat{\epsilon}}_\text{partial} = \textbf{y} - \sum_{i \neq j} {\textbf{x}_i \hat{\boldsymbol{\beta}}_i} \]
calc_partial_resids(clmap_final, xval_inds, 'NUMERIC_AGE', 5) %>%
ggplot(aes(x = x, y = y)) +
geom_point(alpha = .1, size = 1) +
geom_hline(yintercept = 2, color = 'darkgray') +
geom_line(aes(y = y.hat), color = 'red') +
labs(title = 'Numeric age spline versus campaign giving', x = 'age', y = bquote(log[10] ~ 'campaign giving'))
calc_partial_resids(clmap_final, xval_inds, 'NUMERIC_AGE', 5) %>%
ggplot(aes(x = x, y = y.partial.resid)) +
geom_point(alpha = .1, size = 1) +
geom_hline(yintercept = 0, color = 'darkgray') +
geom_line(aes(y = y.hat.partial), color = 'red') +
scale_y_continuous(breaks = seq(-10, 10, by = 2)) +
labs(title = 'Numeric age spline versus partial residuals for campaign giving'
, x = 'age'
, y = 'partial residuals')
Recall that in the PG indicators model, age had a negative coefficient. This paints a more nuanced picture. The first plot regresses campaign giving on the natural spline of age and shows a dip in expected giving for 60-year-olds, a gradual increase to age 80 or so, and then a slow decline. However, we know from previous experience that age is closely correlated with many other predictors of giving (years of giving, total giving, capacity evaluation, and so on). The partial residual plot corrects for this by removing the effect of all the other predictors and regressing the resulting residuals on age. Now, controlling for the other variables, expected campaign giving appears to reach its maximum in the late 40s or early 50s, and decreases thereafter.
Thoughts
Compare the coefficients for both models.
full_join(
data.frame(var = coef(clm_final) %>% names, pg.inds.model = coef(clm_final))
, data.frame(var = coef(clmap_final) %>% names, all.predictors.model = coef(clmap_final))
) %>% mutate(
# Factor levels in order we want them to appear in below table
var = factor(var, levels = c('(Intercept)', 'ACTIVE_PROPOSALS', 'AGE', 'ns(NUMERIC_AGE, df = 5)1'
, 'ns(NUMERIC_AGE, df = 5)2', 'ns(NUMERIC_AGE, df = 5)3', 'ns(NUMERIC_AGE, df = 5)4'
, 'ns(NUMERIC_AGE, df = 5)5', 'PRESIDENT_VISIT', 'PM_VISIT_LAST_2_YRS', 'VISITS_5PLUS'
, 'log10plus1(VISIT_COUNT)', 'AF_25K_GIFT', 'GAVE_IN_LAST_3_YRS', 'sqrt(YEARS_OF_GIVING)'
, 'ns(YEARS_OF_GIVING_LAST_3, df = 2)1', 'ns(YEARS_OF_GIVING_LAST_3, df = 2)2', 'MG_250K_PLUS'
, 'TRUSTEE_OR_ADVISORY_BOARD', 'Alumnus', 'DEEP_ENGAGEMENT', 'CHICAGO_HOME', 'SEASON_TICKET_YEARS'
, 'AFFINITY_SCORE', 'MG_PR_MODEL_DESCBottom Tier', 'MG_PR_MODEL_DESCMiddle Tier', 'MG_PR_MODEL_DESCTop Tier'))
) %>% arrange(var)- The intercept for the all predictors model is much closer to 0. The relatively large intercept for the indicators model likely contributes to its odd residuals behavior (recall the bimodal squared errors).
- Age has a nonlinear relationship with giving. The oldest donors are expected to give slightly less.
- The impact of visits is much smaller for the all predictors model, though more visit activity is always associated with greater giving.
- Giving appears to have a comparable impact between the two models, though interpretation is tricky given that the all predictors model splits its impact over more predictors.
- Alumni status leads to less predicted giving in both models! While surprising on the face of it, this actually fits with what I’ve observed in other datasets. Consider that nonalumni are often added to the database only once they’ve already made a gift, demonstrating both affinity and philanthropic interest.
- The affinity score and MG prioritization models work!
Overall, with Campaign giving as \(y\), several of the PG model variables do seem to offer predictive power above and beyond whatever factors are rolled into the affinity score and MG prioritization models. In particular, age, PM visits, total visits, $25K+ AF gifts, total years of giving, MG gifts, alumni status, and season tickets were still statistically significant when including the other modeled scores. It may be worthwhile paying special attention to these characteristics.
Transaction linear models
Repeat the above, but let \(y_i\) be transformed largest gift or payment.
Cultivation score predictors
Full summary in the appendix.
# List to store transaction linear models
tlms <- list()
for(i in 1:length(xval_inds)) {
# Create linear model excluding the holdout and out-of-sample indices
tlms[[i]] <- mdat %>%
filter(rownum %in% unlist(xval_inds)[unlist(xval_inds) %nin% xval_inds[[i]]]) %>%
lm(
log10plus1(LARGEST_GIFT_OR_PAYMENT) ~
ACTIVE_PROPOSALS +
AGE +
PM_VISIT_LAST_2_YRS +
VISITS_5PLUS +
AF_25K_GIFT +
GAVE_IN_LAST_3_YRS +
MG_250K_PLUS +
PRESIDENT_VISIT +
TRUSTEE_OR_ADVISORY_BOARD +
Alumnus +
DEEP_ENGAGEMENT +
CHICAGO_HOME
, data = .
)
}plot_r2(tlms) +
geom_text(y = seq(10, 200, length.out = k - 1), label = 1:(k - 1), color = 'blue') +
xlim(c(.325, .335))
Here, the mean \(r^2 =\) 0.331 is a bit lower than that seen in the campaign models.
plot_coefs(tlms)
coef_pm_table(tlms, p.sig)Interestingly, all the coefficients are positive besides alumni status, which is not predictive! President visits are hit or miss, which follows given that the quality of visit data is time dependent while gift data is not. The rest all have positive coefficients.
plot_mses(tlms, xval_inds, 'LARGEST_GIFT_OR_PAYMENT')
This is actually much lower than what was seen in the campaign cultivation score model.
plot_resids(tlms, xval_inds, 'LARGEST_GIFT_OR_PAYMENT')$insample
plot_resids(tlms, xval_inds, 'LARGEST_GIFT_OR_PAYMENT')$outsample
This exhibits a similar downward trend as scene in the campaign cultivation score model, so I don’t expect the Q-Q plots to be acceptable.
plot_qq(tlms, xval_inds, 'LARGEST_GIFT_OR_PAYMENT')$insample
plot_qq(tlms, xval_inds, 'LARGEST_GIFT_OR_PAYMENT')$outsample
That looks almost like a discontinuity at -1.
mdat %>% mutate(x.trans = log10plus1(LARGEST_GIFT_OR_PAYMENT)) %>%
ggplot(aes(x = x.trans)) +
geom_histogram(alpha = .5, binwidth = .1) +
geom_density(aes(y = ..count..)) +
geom_vline(aes(xintercept = mean(x.trans)), color = 'blue', linetype = 'dashed') +
geom_vline(aes(xintercept = mean(x.trans) - sd(x.trans)), color = 'blue', linetype = 'dotted')
As before, the combination of binary indicators and all those 0 observations seem to do a number on the model.
plot_qq(tlms, xval_inds, 'LARGEST_GIFT_OR_PAYMENT', filter = 'truth > 0')$insample
plot_qq(tlms, xval_inds, 'LARGEST_GIFT_OR_PAYMENT', filter = 'truth > 0')$outsample
But once again, the expected value conditioned on being a donor is nearly normal. This approach, \(E(\text{giving} ~ | ~ \text{donor status} = 1)\), is definitely worth considering in the future.
All predictors 1
Repeat the above with the underlying variables. Again, the first model will include all variables and I’ll prune it back according to MSE and variable significance. Spline df are set as in the corresponding campaign model.
Full summary in the appendix.
# List to store transaction linear models
tlmaps <- list()
for(i in 1:length(xval_inds)) {
# Create linear model excluding the holdout and out-of-sample indices
tlmaps[[i]] <- mdat %>%
filter(rownum %in% unlist(xval_inds)[unlist(xval_inds) %nin% xval_inds[[i]]]) %>%
lm(
log10plus1(LARGEST_GIFT_OR_PAYMENT) ~
ACTIVE_PROPOSALS +
ns(NUMERIC_AGE, df = 5) + # Underlying variable to AGE indicator
PM_VISIT_LAST_2_YRS +
log10plus1(VISIT_COUNT) + # Underlying VISITS_5PLUS indicator
AF_25K_GIFT +
sqrt(YEARS_OF_GIVING) + # Underlying GAVE_IN_LAST_3_YRS
ns(YEARS_OF_GIVING_LAST_3, df = 2) + # Underlying GAVE_IN_LAST_3_YRS
MG_250K_PLUS + # Decided to leave as factor
PRESIDENT_VISIT +
TRUSTEE_OR_ADVISORY_BOARD +
Alumnus +
DOUBLE_ALUM + # Deep Engagement component
EVER_PARENT + # Deep Engagement component
SEASON_TICKET_YEARS + # Deep Engagement component
CHICAGO_HOME +
QUAL_LEVEL +
AFFINITY_SCORE +
MG_PR_MODEL_DESC
, data = .
)
}plot_r2(tlmaps) +
geom_text(y = seq(10, 250, length.out = k - 1), label = 1:(k-1), color = 'blue') +
xlim(c(.593, .605))
plot_mses(tlmaps, xval_inds, 'LARGEST_GIFT_OR_PAYMENT')
While \(r^2\) is not as high as that achieved by the campaign all predictors model, this is the lowest MSE seen yet.
plot_coefs(tlmaps)
coef_pm_table(tlmaps, p.sig)Intersetingly, there are some notable differences compared to campaign giving. Here, having a Chicago home address does matter, as do parent affinity, and committee participation. Qualification level is still inconsistent, and age appears to matter much less here compared to the campaign model. Interestingly, compared to the previous model, alumni status is significant again once these underlying factors are included.
plot_resids(tlmaps, xval_inds, 'LARGEST_GIFT_OR_PAYMENT')$outsample
plot_qq(tlmaps, xval_inds, 'LARGEST_GIFT_OR_PAYMENT')$outsample
That residuals plot is still quite poor, but the Q-Q plot is respectable.
All predictors 2
Re-fit the above, dropping qualification level, active proposals, and double alum. It also appears that fewer splie df for age are called for.
Full results in the appendix.
# List to store transaction linear models
tlmaps2 <- list()
for(i in 1:length(xval_inds)) {
# Create linear model excluding the holdout and out-of-sample indices
tlmaps2[[i]] <- mdat %>%
filter(rownum %in% unlist(xval_inds)[unlist(xval_inds) %nin% xval_inds[[i]]]) %>%
lm(
log10plus1(LARGEST_GIFT_OR_PAYMENT) ~
ns(NUMERIC_AGE, df = 3) + # Underlying variable to AGE indicator
PM_VISIT_LAST_2_YRS +
log10plus1(VISIT_COUNT) + # Underlying VISITS_5PLUS indicator
AF_25K_GIFT +
sqrt(YEARS_OF_GIVING) + # Underlying GAVE_IN_LAST_3_YRS
ns(YEARS_OF_GIVING_LAST_3, df = 2) + # Underlying GAVE_IN_LAST_3_YRS
MG_250K_PLUS + # Decided to leave as factor
PRESIDENT_VISIT +
TRUSTEE_OR_ADVISORY_BOARD +
Alumnus +
EVER_PARENT + # Deep Engagement component
SEASON_TICKET_YEARS + # Deep Engagement component
CHICAGO_HOME +
AFFINITY_SCORE +
MG_PR_MODEL_DESC
, data = .
)
}plot_r2(tlmaps2) +
geom_text(y = seq(10, 250, length.out = k - 1), label = 1:(k-1), color = 'blue') +
xlim(c(.59, .60))
plot_mses(tlmaps2, xval_inds, 'LARGEST_GIFT_OR_PAYMENT')
Similar \(r^2\) and error, so the smaller model is preferred.
plot_coefs(tlmaps2)
coef_pm_table(tlmaps2, p.sig)Alumni and parent affiliation reduce the expected largest gift amount, on average.
plot_resids(tlmaps2, xval_inds, 'LARGEST_GIFT_OR_PAYMENT')$outsample
plot_qq(tlmaps2, xval_inds, 'LARGEST_GIFT_OR_PAYMENT')$outsample
Still can’t seem to fix those residuals.
Comparison
Check the models on out-of-sample data. Full results in the appendix.
# Predict largest transaction, PG score indicators
tlm_final <- tlms[[1]] %>% update(data = mdat %>% filter(rownum %in% unlist(xval_inds)))
# Predict largest transaction, underlying factors
tlmap_final <- tlmaps2[[1]] %>% update(data = mdat %>% filter(rownum %in% unlist(xval_inds)))plot_resids(list(tlmap_final, tlm_final), list(oos_inds, oos_inds), 'LARGEST_GIFT_OR_PAYMENT')$outsample +
scale_x_continuous(breaks = seq(0, 10, by = 2)) +
scale_y_continuous(breaks = seq(-10, 10, by = 2)) +
scale_color_discrete(labels = c('All predictors', 'PG indicators')) +
labs(color = 'Model')
Honestly, neither model looks great.
plot_qq(list(tlmap_final, tlm_final), list(oos_inds, oos_inds), 'LARGEST_GIFT_OR_PAYMENT')$outsample +
scale_y_continuous(breaks = seq(-10, 10, by = 2)) +
scale_color_discrete(labels = c('All predictors', 'PG indicators')) +
labs(color = 'Model')
At least the the all predictors residuals lie closer to a normal distribution.
t_mses_final <- rbind(
calc_preds(list(tlm_final), list(oos_inds), 'LARGEST_GIFT_OR_PAYMENT')[[1]] %>%
mutate(model = 'PG indicators')
, calc_preds(list(tlmap_final), list(oos_inds), 'LARGEST_GIFT_OR_PAYMENT')[[1]] %>%
mutate(model = 'All predictors')
) %>% mutate(
model = factor(model)
, sq.error = (truth - preds)^2
)
t_mses_final <- t_mses_final %>% left_join(
t_mses_final %>% group_by(model) %>% summarise(mse = mean(sq.error))
, by = c('model', 'model')
)
t_mses_final %>%
ggplot(aes(x = sq.error, color = model)) +
geom_density() +
geom_vline(aes(xintercept = mean(sq.error)), linetype = 'dotted', alpha = .5) +
geom_vline(aes(xintercept = mse, color = model), linetype = 'dashed') +
xlim(c(0, 5)) +
labs(
x = 'squared error'
, title = bquote('Squared error across models, means = ' ~
.(t_mses_final %>% group_by(model) %>% summarise(mse = mean(mse)) %>% select(mse) %>%
unlist() %>% round(3) %>% paste(collapse = ', ')))
)
Also very close to the previous results. Again, all predictors is a better regression model.
Finally, consider the effect of age on largest gift.
calc_partial_resids(tlmap_final, xval_inds, 'NUMERIC_AGE', 3) %>%
ggplot(aes(x = x, y = y)) +
geom_point(alpha = .1, size = 1) +
geom_hline(yintercept = 2, color = 'darkgray') +
geom_line(aes(y = y.hat), color = 'red') +
labs(title = 'Numeric age spline versus largest transaction', x = 'age', y = bquote(log[10] ~ 'largest transaction'))
calc_partial_resids(tlmap_final, xval_inds, 'NUMERIC_AGE', 3) %>%
ggplot(aes(x = x, y = y.partial.resid)) +
geom_point(alpha = .1, size = 1) +
geom_hline(yintercept = 0, color = 'darkgray') +
geom_line(aes(y = y.hat.partial), color = 'red') +
labs(title = 'Numeric age spline versus partial residuals for largest transaction'
, x = 'age'
, y = 'partial residuals')
I find this surprising. Unlike in the campaign model comparison, largest gift size appears to increase monotonically with age. However, controlling for the other variables, we again see peak giving occurs in the late 40s or early 50s, though the fitted line is nearly flat (confirming the lesser impact of age in this model compared to the previous one). Apparently there are very few large estate gifts compared to the number of older donors.
Takeaway. Today’s age is not particularly predictive of largest gift; in future models consider age at time of gift instead. This could also be used to predict next gift size.
Thoughts
Here are the coefficients for each model.
full_join(
data.frame(var = coef(tlm_final) %>% names, pg.inds.model = coef(tlm_final))
, data.frame(var = coef(tlmap_final) %>% names, all.predictors.model = coef(tlmap_final))
) %>% mutate(
# Factor levels in order we want them to appear in below table
var = factor(var, levels = c('(Intercept)', 'ACTIVE_PROPOSALS'
, 'AGE', 'ns(NUMERIC_AGE, df = 3)1', 'ns(NUMERIC_AGE, df = 3)2', 'ns(NUMERIC_AGE, df = 3)3'
, 'PRESIDENT_VISIT', 'PM_VISIT_LAST_2_YRS', 'VISITS_5PLUS', 'log10plus1(VISIT_COUNT)'
, 'AF_25K_GIFT', 'GAVE_IN_LAST_3_YRS', 'sqrt(YEARS_OF_GIVING)'
, 'ns(YEARS_OF_GIVING_LAST_3, df = 2)1', 'ns(YEARS_OF_GIVING_LAST_3, df = 2)2', 'MG_250K_PLUS'
, 'TRUSTEE_OR_ADVISORY_BOARD'
, 'Alumnus', 'EVER_PARENT', 'DEEP_ENGAGEMENT', 'CHICAGO_HOME', 'SEASON_TICKET_YEARS'
, 'AFFINITY_SCORE', 'MG_PR_MODEL_DESCBottom Tier', 'MG_PR_MODEL_DESCMiddle Tier', 'MG_PR_MODEL_DESCTop Tier'))
) %>% arrange(var)- The intercept is smaller in the all predictors model.
- Controlling for the other predictors, the effect of age is fairly minor.
- Unsurprisingly, past major giving is the strongest predictor of…major gift behavior.
- Again, the affinity score and prioritization models work – though here, bottom and middle tier don’t appear to have given at different levels in the past (which is likely why they weren’t top tier).
All of the PG indicators apparently help estimate largest gift amount, above and beyond the information that other variables and modeled scores provide. Importantly, with the exception of alumni status, all the signs point in the same direction.
Final Comparison
Rebuild all models using the entire dataset. Full results are in the appendix.
# Final models list
final_models <- list(
campaign.pg.score = clm_final %>% update(data = mdat %>% filter(rownum %nin% outliers))
, largest.trans.pg.score = tlm_final %>% update(data = mdat %>% filter(rownum %nin% outliers))
, campaign.all.preds = clmap_final %>% update(data = mdat %>% filter(rownum %nin% outliers))
, largest.trans.all.preds = tlmap_final %>% update(data = mdat %>% filter(rownum %nin% outliers))
)
# Coefficients data frame
final_coefs <- full_join(
data.frame(var = coef(final_models[[1]]) %>% names()
, campaign.pg = coef(final_models[[1]]))
, data.frame(var = coef(final_models[[2]]) %>% names()
, largest.pg = coef(final_models[[2]]))
) %>% full_join(
data.frame(var = coef(final_models[[3]]) %>% names() %>% str_replace(', df = [0-9]*', '')
, campaign.all = coef(final_models[[3]]))
) %>% full_join(
data.frame(var = coef(final_models[[4]]) %>% names() %>% str_replace(', df = [0-9]*', '')
, largest.all = coef(final_models[[4]]))
) %>% mutate(
# Factor levels in order we want them to appear in below table
var = factor(var, levels = c('(Intercept)', 'ACTIVE_PROPOSALS'
, 'AGE', 'ns(NUMERIC_AGE)1', 'ns(NUMERIC_AGE)2', 'ns(NUMERIC_AGE)3', 'ns(NUMERIC_AGE)4', 'ns(NUMERIC_AGE)5'
, 'PRESIDENT_VISIT', 'PM_VISIT_LAST_2_YRS', 'VISITS_5PLUS', 'log10plus1(VISIT_COUNT)'
, 'AF_25K_GIFT', 'GAVE_IN_LAST_3_YRS', 'sqrt(YEARS_OF_GIVING)'
, 'ns(YEARS_OF_GIVING_LAST_3)1', 'ns(YEARS_OF_GIVING_LAST_3)2', 'MG_250K_PLUS'
, 'TRUSTEE_OR_ADVISORY_BOARD'
, 'Alumnus', 'EVER_PARENT', 'DEEP_ENGAGEMENT', 'CHICAGO_HOME', 'ns(SEASON_TICKET_YEARS)'
, 'AFFINITY_SCORE', 'MG_PR_MODEL_DESCBottom Tier', 'MG_PR_MODEL_DESCMiddle Tier', 'MG_PR_MODEL_DESCTop Tier'))
) %>% arrange(var)final_coefs %>% mutate_if(is.numeric, function(x) round(x, 3))- Unsurprisingly, all intercepts are positive. The campaign all predictors model is closest to 0, and the all predictor models have lower intercepts across the board.
- Intrestingly, alumni status is negative for all but the largest gift regressed on PG checklist model (which was barely significant).
Let’s normalize these using the MG_250K_PLUS indicator appearing in all four models.
final_coefs %>%
mutate_if(is.numeric, function(x) {x/x[which(final_coefs$var == 'MG_250K_PLUS')]} %>% round(4)) %>%
gather('model', 'coefficient', 2:5) %>%
mutate(model = factor(model, levels = c('campaign.pg', 'largest.pg', 'campaign.all', 'largest.all'))) %>%
ggplot(aes(x = var, y = coefficient, fill = model)) +
geom_col(position = position_dodge(), width = .7) +
geom_hline(yintercept = 0, color = 'darkgray') +
scale_y_continuous(limits = c(-1, 3), breaks = seq(-1, 3, by = .5)) +
theme(
axis.text.x = element_text(angle = 90, hjust = 1, vjust = .5)
, axis.title.y = element_text(angle = 0, vjust = .5)
) +
labs(title = 'Model coefficient comparison, scaled to MG = 1', y = bquote(hat(beta)))
- When predicting campaign giving, giving recency (gave in last 3) is more useful than previous MG behavior! Note that loyalty is also a much better predictor than the MG PR model.
- Visits are also very impactful on average.
- Most of the coefficients shared between models point in the same direction, with the notable exceptions of age and MG model tiers.
The exact effect of age and recent years of giving are difficult to see since multiple coefficients are computed per spline. Consider the partial residual plots for each model.
final_partials <- rbind(
data.frame(
model = 'campaign ~ pg score'
, calc_partial_resids(final_models$campaign.pg.score, inds = c(xval_inds, oos_inds), xname = 'AGE')
) %>% mutate(x = mdat %>% filter(rownum %in% unlist(c(xval_inds, oos_inds))) %>% select(NUMERIC_AGE) %>% unlist())
, data.frame(
model = 'largest trans ~ pg score'
, calc_partial_resids(final_models$largest.trans.pg.score, inds = c(xval_inds, oos_inds), xname = 'AGE')
) %>% mutate(x = mdat %>% filter(rownum %in% unlist(c(xval_inds, oos_inds))) %>% select(NUMERIC_AGE) %>% unlist())
, data.frame(
model = 'campaign ~ all preds'
, calc_partial_resids(final_models$campaign.all.preds, inds = c(xval_inds, oos_inds)
, xname = 'NUMERIC_AGE', df = 5)
)
, data.frame(
model = 'largest trans ~ all preds'
, calc_partial_resids(final_models$largest.trans.all.preds, inds = c(xval_inds, oos_inds)
, xname = 'NUMERIC_AGE', df = 3)
)
)final_partials %>%
ggplot(aes(x = x, y = y)) +
geom_point(alpha = .1, size = 1) +
geom_hline(yintercept = 2, color = 'darkgray') +
geom_line(aes(y = y.hat, color = model), size = 1, alpha = .8) +
labs(title = 'Numeric age spline versus giving', x = 'age', y = bquote(log[10] ~ 'giving'))
final_partials %>%
ggplot(aes(x = x, y = y.partial.resid)) +
geom_point(alpha = .1, size = 1) +
geom_hline(yintercept = 0, color = 'darkgray') +
geom_line(aes(y = y.hat.partial, color = model), size = 1, alpha = .8) +
labs(title = 'Numeric age spline versus partial residuals for giving'
, x = 'age'
, y = 'partial residuals')
- In all four models, age has only a small impact after controlling for the other predictors.
- The nonlinear splines fits suggest that peak giving occurs in the late 40s or early 50s, and the age 60+ indicator does not provide information beyond that included in other predictors.
final_partials <- rbind(
data.frame(
model = 'campaign ~ pg score'
, calc_partial_resids(final_models$campaign.pg.score, inds = c(xval_inds, oos_inds)
, xname = 'GAVE_IN_LAST_3_YRS')
)
, data.frame(
model = 'largest trans ~ pg score'
, calc_partial_resids(final_models$largest.trans.pg.score, inds = c(xval_inds, oos_inds)
, xname = 'GAVE_IN_LAST_3_YRS')
)
, data.frame(
model = 'campaign ~ all preds'
, calc_partial_resids(final_models$campaign.all.preds, inds = c(xval_inds, oos_inds)
, xname = 'YEARS_OF_GIVING_LAST_3', df = 2)
)
, data.frame(
model = 'largest trans ~ all preds'
, calc_partial_resids(final_models$largest.trans.all.preds, inds = c(xval_inds, oos_inds)
, xname = 'YEARS_OF_GIVING_LAST_3', df = 2)
)
)final_partials %>%
ggplot(aes(x = x, y = y)) +
geom_point(alpha = .1, size = 1) +
geom_hline(yintercept = 2, color = 'darkgray') +
geom_line(aes(y = y.hat, color = model), size = 1, alpha = .8) +
labs(title = 'Years of giving last 3 versus giving', x = 'years', y = bquote(log[10] ~ 'giving'))
final_partials %>%
ggplot(aes(x = x, y = y.partial.resid)) +
geom_point(alpha = .1, size = 1) +
geom_hline(yintercept = 0, color = 'darkgray') +
geom_line(aes(y = y.hat.partial, color = model, group = model), size = 1, alpha = .8) +
scale_y_continuous(breaks = -10:10) +
labs(title = 'Years of giving last 3 versus partial residuals for giving'
, x = 'years'
, y = 'partial residuals')
rbind(
data.frame(
model = 'campaign ~ all preds'
, x = mdat %>% filter(rownum %in% unlist(c(xval_inds, oos_inds))) %>% select(YEARS_OF_GIVING) %>% unlist()
, y = {final_models$campaign.all.preds %>% model.frame()}[, 1]
, y.partial.resid = {final_models$campaign.all.preds %>% residuals(type = 'partial')}[, 'sqrt(YEARS_OF_GIVING)']
) %>% mutate(y.hat.partial = lm(y.partial.resid ~ sqrt(x)) %>% fitted())
, data.frame(
model = 'largest trans ~ all preds'
, x = mdat %>% filter(rownum %in% unlist(c(xval_inds, oos_inds))) %>% select(YEARS_OF_GIVING) %>% unlist()
, y = {final_models$largest.trans.all.preds %>% model.frame()}[, 1]
, y.partial.resid = {final_models$largest.trans.all.preds %>% residuals(type = 'partial')}[, 'sqrt(YEARS_OF_GIVING)']
) %>% mutate(y.hat.partial = lm(y.partial.resid ~ sqrt(x)) %>% fitted())
) %>%
ggplot(aes(x = x, y = y.partial.resid, color = model)) +
geom_point(alpha = .1, size = 1) +
geom_hline(yintercept = 0, color = 'darkgray') +
geom_line(aes(y = y.hat.partial, group = model), size = 1, alpha = .8) +
scale_y_continuous(breaks = -10:10) +
labs(title = 'Years of giving versus partial residuals for giving'
, x = 'years of giving'
, y = 'partial residuals')
- The PG score models use a binary indicator requiring at least one gift in the last three years, and at least 10 years of giving total, while the all predictors models separate these two components and treat them numerically.
- Controlling for the effects of the other predictors, recency of giving has a large impact on overall campaign giving in the all predictors model.
- Looking at years of giving, interestingly both lines cross 0 before 10, suggesting a smaller cutoff (7 or 8 years?) could be used in the PG checklist.
- When modeling expected campaign giving it’s certainly better to treat these variables separately. Largest transaction appears to have a less clear relationship (as seen in the above plots).
Conclusions
The goal of this project was to look at the impact of each of the PG checklist items has had on giving, and to see whether I could build a reasonable regression model based on the checklist.
The checklist succeeds at its primary goal of identifying factors impacting giving transaction amounts – in the largest transaction regressed on checklist items model, while alumni status is not significant and president visits are marginal, all the factors are positively associated with giving. A past $250K+ MG is by far the most important factor, with $25K+ annual giving, loyalty (gave in last 3 years), and 5+ visits approximately tied for second place. Unfortunately, modeling on largest gift consistently overestimates giving amounts at the high end, and the other variables I examined weren’t able to correct for this. The modeled results could still be interesting for prioritization purposes, though.
The big winner was the campaign giving regressed on all predictors model, with nice-looking residual plots exhibiting only a bit of positive skewness. Interestingly, the best predictor by far is donor loyalty, measured by the number of years of giving in the last three, nearly doubling the impact of having made a previous $250K+ gift!
I also included previous modeled scores (affinity score) and tiering (MG prioritization) in the all predictors models to see whether the PG checklist items add information beyond that contained in these earlier models. Deep engagement, Chicago home address, committee participation, and president visits fall out, but the others are still impactful. Interestingly, alumni status and age don’t matter as much as we usually think (alumni status hurts, and age has a nonmonotonic association with giving).
Some ideas for improvements:
Evaluation rating significantly impacts giving at some levels but not others, due partially to small sample size but also probably due to binning. Can a continuous rating prediction be developed?
Is age a biased predictor? Perhaps age at the time of the largest gift should be used instead of today’s age.
Consider a regression model conditional on making a gift, i.e.
\[ E(\text{giving amount} ~ | ~ \text{donor status} = 1) \]
- Consider a time series approach, where future giving over some reasonable cultivation period (2 years?) is estimated conditional on previous data.
\[ E \left( (\text{giving amount})_t ~ | ~ (\text{donor attributes})_{t - 1} \right) \]
Appendix
Campaign cultivation results
lapply(clms, function(x) summary(x))[[1]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ACTIVE_PROPOSALS +
AGE + PM_VISIT_LAST_2_YRS + VISITS_5PLUS + AF_25K_GIFT +
GAVE_IN_LAST_3_YRS + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DEEP_ENGAGEMENT + CHICAGO_HOME, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.8239 -1.2279 -0.1471 0.9867 5.6858
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.34693 0.02810 47.941 < 2e-16 ***
ACTIVE_PROPOSALS 0.29446 0.04558 6.460 1.07e-10 ***
AGE -0.21472 0.02142 -10.022 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.59623 0.04824 12.361 < 2e-16 ***
VISITS_5PLUS 0.75279 0.03485 21.598 < 2e-16 ***
AF_25K_GIFT 0.85477 0.06019 14.200 < 2e-16 ***
GAVE_IN_LAST_3_YRS 2.03758 0.02332 87.385 < 2e-16 ***
MG_250K_PLUS 1.12287 0.10212 10.995 < 2e-16 ***
PRESIDENT_VISIT 0.42916 0.08478 5.062 4.19e-07 ***
TRUSTEE_OR_ADVISORY_BOARD 0.26704 0.04512 5.918 3.31e-09 ***
Alumnus -0.11905 0.02771 -4.296 1.74e-05 ***
DEEP_ENGAGEMENT 0.32417 0.02300 14.093 < 2e-16 ***
CHICAGO_HOME 0.10052 0.02231 4.505 6.66e-06 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 1.344 on 18765 degrees of freedom
Multiple R-squared: 0.4662, Adjusted R-squared: 0.4659
F-statistic: 1366 on 12 and 18765 DF, p-value: < 2.2e-16
[[2]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ACTIVE_PROPOSALS +
AGE + PM_VISIT_LAST_2_YRS + VISITS_5PLUS + AF_25K_GIFT +
GAVE_IN_LAST_3_YRS + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DEEP_ENGAGEMENT + CHICAGO_HOME, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.0307 -1.2214 -0.1514 0.9819 5.4201
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.33066 0.02805 47.436 < 2e-16 ***
ACTIVE_PROPOSALS 0.30700 0.04577 6.708 2.03e-11 ***
AGE -0.21829 0.02139 -10.204 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.59216 0.04831 12.259 < 2e-16 ***
VISITS_5PLUS 0.75868 0.03481 21.793 < 2e-16 ***
AF_25K_GIFT 0.79731 0.06102 13.067 < 2e-16 ***
GAVE_IN_LAST_3_YRS 2.03779 0.02328 87.533 < 2e-16 ***
MG_250K_PLUS 1.17113 0.09934 11.789 < 2e-16 ***
PRESIDENT_VISIT 0.45526 0.08439 5.395 6.94e-08 ***
TRUSTEE_OR_ADVISORY_BOARD 0.28420 0.04532 6.270 3.68e-10 ***
Alumnus -0.10930 0.02767 -3.950 7.84e-05 ***
DEEP_ENGAGEMENT 0.34208 0.02290 14.940 < 2e-16 ***
CHICAGO_HOME 0.08937 0.02228 4.012 6.05e-05 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 1.34 on 18765 degrees of freedom
Multiple R-squared: 0.4696, Adjusted R-squared: 0.4693
F-statistic: 1385 on 12 and 18765 DF, p-value: < 2.2e-16
[[3]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ACTIVE_PROPOSALS +
AGE + PM_VISIT_LAST_2_YRS + VISITS_5PLUS + AF_25K_GIFT +
GAVE_IN_LAST_3_YRS + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DEEP_ENGAGEMENT + CHICAGO_HOME, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.8276 -1.2195 -0.1510 0.9866 5.7040
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.33938 0.02800 47.838 < 2e-16 ***
ACTIVE_PROPOSALS 0.29632 0.04583 6.465 1.04e-10 ***
AGE -0.22454 0.02146 -10.465 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.61762 0.04842 12.754 < 2e-16 ***
VISITS_5PLUS 0.72950 0.03477 20.982 < 2e-16 ***
AF_25K_GIFT 0.84794 0.06027 14.070 < 2e-16 ***
GAVE_IN_LAST_3_YRS 2.04916 0.02337 87.684 < 2e-16 ***
MG_250K_PLUS 1.08151 0.10098 10.710 < 2e-16 ***
PRESIDENT_VISIT 0.45119 0.08407 5.367 8.12e-08 ***
TRUSTEE_OR_ADVISORY_BOARD 0.26498 0.04551 5.822 5.90e-09 ***
Alumnus -0.11991 0.02762 -4.342 1.42e-05 ***
DEEP_ENGAGEMENT 0.33789 0.02294 14.731 < 2e-16 ***
CHICAGO_HOME 0.09713 0.02236 4.344 1.41e-05 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 1.343 on 18765 degrees of freedom
Multiple R-squared: 0.4676, Adjusted R-squared: 0.4673
F-statistic: 1374 on 12 and 18765 DF, p-value: < 2.2e-16
[[4]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ACTIVE_PROPOSALS +
AGE + PM_VISIT_LAST_2_YRS + VISITS_5PLUS + AF_25K_GIFT +
GAVE_IN_LAST_3_YRS + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DEEP_ENGAGEMENT + CHICAGO_HOME, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.8414 -1.2329 -0.1434 0.9886 5.6898
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.36002 0.02802 48.532 < 2e-16 ***
ACTIVE_PROPOSALS 0.28005 0.04595 6.095 1.12e-09 ***
AGE -0.22377 0.02143 -10.440 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.61631 0.04860 12.680 < 2e-16 ***
VISITS_5PLUS 0.74954 0.03475 21.568 < 2e-16 ***
AF_25K_GIFT 0.87792 0.06035 14.547 < 2e-16 ***
GAVE_IN_LAST_3_YRS 2.03790 0.02332 87.375 < 2e-16 ***
MG_250K_PLUS 1.08349 0.10146 10.679 < 2e-16 ***
PRESIDENT_VISIT 0.46548 0.08481 5.488 4.11e-08 ***
TRUSTEE_OR_ADVISORY_BOARD 0.26591 0.04563 5.828 5.70e-09 ***
Alumnus -0.12710 0.02764 -4.598 4.29e-06 ***
DEEP_ENGAGEMENT 0.31337 0.02292 13.674 < 2e-16 ***
CHICAGO_HOME 0.09791 0.02231 4.388 1.15e-05 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 1.344 on 18765 degrees of freedom
Multiple R-squared: 0.4654, Adjusted R-squared: 0.4651
F-statistic: 1362 on 12 and 18765 DF, p-value: < 2.2e-16
[[5]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ACTIVE_PROPOSALS +
AGE + PM_VISIT_LAST_2_YRS + VISITS_5PLUS + AF_25K_GIFT +
GAVE_IN_LAST_3_YRS + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DEEP_ENGAGEMENT + CHICAGO_HOME, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.8844 -1.2314 -0.1522 0.9870 5.6900
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.35196 0.02817 47.984 < 2e-16 ***
ACTIVE_PROPOSALS 0.26092 0.04625 5.641 1.71e-08 ***
AGE -0.22244 0.02151 -10.339 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.64613 0.04867 13.276 < 2e-16 ***
VISITS_5PLUS 0.75115 0.03509 21.406 < 2e-16 ***
AF_25K_GIFT 0.83049 0.06063 13.699 < 2e-16 ***
GAVE_IN_LAST_3_YRS 2.03688 0.02345 86.864 < 2e-16 ***
MG_250K_PLUS 1.08036 0.10156 10.638 < 2e-16 ***
PRESIDENT_VISIT 0.48657 0.08422 5.777 7.70e-09 ***
TRUSTEE_OR_ADVISORY_BOARD 0.26540 0.04542 5.843 5.21e-09 ***
Alumnus -0.12056 0.02779 -4.339 1.44e-05 ***
DEEP_ENGAGEMENT 0.33543 0.02308 14.536 < 2e-16 ***
CHICAGO_HOME 0.09086 0.02238 4.060 4.94e-05 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 1.347 on 18765 degrees of freedom
Multiple R-squared: 0.4634, Adjusted R-squared: 0.4631
F-statistic: 1351 on 12 and 18765 DF, p-value: < 2.2e-16
[[6]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ACTIVE_PROPOSALS +
AGE + PM_VISIT_LAST_2_YRS + VISITS_5PLUS + AF_25K_GIFT +
GAVE_IN_LAST_3_YRS + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DEEP_ENGAGEMENT + CHICAGO_HOME, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.8287 -1.2295 -0.1432 0.9848 5.7026
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.35238 0.02807 48.176 < 2e-16 ***
ACTIVE_PROPOSALS 0.28734 0.04601 6.245 4.34e-10 ***
AGE -0.23307 0.02143 -10.877 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.59596 0.04873 12.230 < 2e-16 ***
VISITS_5PLUS 0.75457 0.03483 21.665 < 2e-16 ***
AF_25K_GIFT 0.84636 0.06145 13.774 < 2e-16 ***
GAVE_IN_LAST_3_YRS 2.04433 0.02338 87.450 < 2e-16 ***
MG_250K_PLUS 1.16030 0.10296 11.269 < 2e-16 ***
PRESIDENT_VISIT 0.43727 0.08545 5.117 3.13e-07 ***
TRUSTEE_OR_ADVISORY_BOARD 0.26291 0.04535 5.798 6.83e-09 ***
Alumnus -0.12289 0.02762 -4.449 8.70e-06 ***
DEEP_ENGAGEMENT 0.33319 0.02297 14.506 < 2e-16 ***
CHICAGO_HOME 0.10104 0.02229 4.534 5.83e-06 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 1.344 on 18765 degrees of freedom
Multiple R-squared: 0.4645, Adjusted R-squared: 0.4642
F-statistic: 1356 on 12 and 18765 DF, p-value: < 2.2e-16
[[7]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ACTIVE_PROPOSALS +
AGE + PM_VISIT_LAST_2_YRS + VISITS_5PLUS + AF_25K_GIFT +
GAVE_IN_LAST_3_YRS + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DEEP_ENGAGEMENT + CHICAGO_HOME, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.8764 -1.2307 -0.1563 0.9935 5.6954
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.33071 0.02823 47.131 < 2e-16 ***
ACTIVE_PROPOSALS 0.23338 0.04607 5.066 4.09e-07 ***
AGE -0.22714 0.02149 -10.571 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.63896 0.04846 13.185 < 2e-16 ***
VISITS_5PLUS 0.75737 0.03482 21.754 < 2e-16 ***
AF_25K_GIFT 0.84555 0.06045 13.987 < 2e-16 ***
GAVE_IN_LAST_3_YRS 2.03324 0.02344 86.744 < 2e-16 ***
MG_250K_PLUS 1.13352 0.10301 11.004 < 2e-16 ***
PRESIDENT_VISIT 0.46217 0.08479 5.451 5.07e-08 ***
TRUSTEE_OR_ADVISORY_BOARD 0.27520 0.04521 6.087 1.17e-09 ***
Alumnus -0.09998 0.02778 -3.599 0.00032 ***
DEEP_ENGAGEMENT 0.34479 0.02311 14.918 < 2e-16 ***
CHICAGO_HOME 0.10005 0.02235 4.476 7.65e-06 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 1.346 on 18765 degrees of freedom
Multiple R-squared: 0.463, Adjusted R-squared: 0.4627
F-statistic: 1348 on 12 and 18765 DF, p-value: < 2.2e-16
[[8]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ACTIVE_PROPOSALS +
AGE + PM_VISIT_LAST_2_YRS + VISITS_5PLUS + AF_25K_GIFT +
GAVE_IN_LAST_3_YRS + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DEEP_ENGAGEMENT + CHICAGO_HOME, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.8964 -1.2328 -0.1415 0.9853 5.6799
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.34196 0.02812 47.718 < 2e-16 ***
ACTIVE_PROPOSALS 0.30133 0.04590 6.566 5.32e-11 ***
AGE -0.21375 0.02145 -9.966 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.59726 0.04869 12.267 < 2e-16 ***
VISITS_5PLUS 0.74967 0.03486 21.507 < 2e-16 ***
AF_25K_GIFT 0.81077 0.06108 13.275 < 2e-16 ***
GAVE_IN_LAST_3_YRS 2.03577 0.02337 87.095 < 2e-16 ***
MG_250K_PLUS 1.11914 0.10216 10.955 < 2e-16 ***
PRESIDENT_VISIT 0.49984 0.08546 5.849 5.03e-09 ***
TRUSTEE_OR_ADVISORY_BOARD 0.26241 0.04594 5.712 1.13e-08 ***
Alumnus -0.10918 0.02772 -3.938 8.24e-05 ***
DEEP_ENGAGEMENT 0.32971 0.02299 14.342 < 2e-16 ***
CHICAGO_HOME 0.09713 0.02237 4.341 1.42e-05 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 1.345 on 18765 degrees of freedom
Multiple R-squared: 0.4632, Adjusted R-squared: 0.4628
F-statistic: 1349 on 12 and 18765 DF, p-value: < 2.2e-16
[[9]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ACTIVE_PROPOSALS +
AGE + PM_VISIT_LAST_2_YRS + VISITS_5PLUS + AF_25K_GIFT +
GAVE_IN_LAST_3_YRS + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DEEP_ENGAGEMENT + CHICAGO_HOME, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.8367 -1.2275 -0.1522 0.9824 5.6898
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.34522 0.02808 47.910 < 2e-16 ***
ACTIVE_PROPOSALS 0.28844 0.04549 6.340 2.34e-10 ***
AGE -0.21832 0.02151 -10.152 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.61760 0.04808 12.845 < 2e-16 ***
VISITS_5PLUS 0.74641 0.03491 21.383 < 2e-16 ***
AF_25K_GIFT 0.79725 0.06070 13.134 < 2e-16 ***
GAVE_IN_LAST_3_YRS 2.04948 0.02343 87.466 < 2e-16 ***
MG_250K_PLUS 1.06994 0.10179 10.511 < 2e-16 ***
PRESIDENT_VISIT 0.44647 0.08442 5.288 1.25e-07 ***
TRUSTEE_OR_ADVISORY_BOARD 0.26236 0.04543 5.775 7.80e-09 ***
Alumnus -0.11769 0.02767 -4.253 2.12e-05 ***
DEEP_ENGAGEMENT 0.32273 0.02304 14.006 < 2e-16 ***
CHICAGO_HOME 0.09442 0.02235 4.224 2.41e-05 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 1.346 on 18765 degrees of freedom
Multiple R-squared: 0.4659, Adjusted R-squared: 0.4655
F-statistic: 1364 on 12 and 18765 DF, p-value: < 2.2e-16
[[10]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ACTIVE_PROPOSALS +
AGE + PM_VISIT_LAST_2_YRS + VISITS_5PLUS + AF_25K_GIFT +
GAVE_IN_LAST_3_YRS + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DEEP_ENGAGEMENT + CHICAGO_HOME, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.8200 -1.2335 -0.1495 0.9899 5.6845
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.34226 0.02813 47.722 < 2e-16 ***
ACTIVE_PROPOSALS 0.28224 0.04581 6.161 7.37e-10 ***
AGE -0.21910 0.02144 -10.221 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.60690 0.04828 12.571 < 2e-16 ***
VISITS_5PLUS 0.74335 0.03466 21.445 < 2e-16 ***
AF_25K_GIFT 0.86533 0.06029 14.353 < 2e-16 ***
GAVE_IN_LAST_3_YRS 2.03961 0.02335 87.365 < 2e-16 ***
MG_250K_PLUS 1.06492 0.10174 10.467 < 2e-16 ***
PRESIDENT_VISIT 0.41305 0.08505 4.857 1.20e-06 ***
TRUSTEE_OR_ADVISORY_BOARD 0.29067 0.04543 6.398 1.62e-10 ***
Alumnus -0.10872 0.02775 -3.918 8.96e-05 ***
DEEP_ENGAGEMENT 0.31888 0.02300 13.867 < 2e-16 ***
CHICAGO_HOME 0.10224 0.02240 4.565 5.03e-06 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 1.345 on 18761 degrees of freedom
Multiple R-squared: 0.465, Adjusted R-squared: 0.4646
F-statistic: 1359 on 12 and 18761 DF, p-value: < 2.2e-16Campaign all predictors 1
lapply(clmaps, function(x) summary(x))[[1]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ACTIVE_PROPOSALS +
ns(NUMERIC_AGE, df = 5) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DOUBLE_ALUM + EVER_PARENT + SEASON_TICKET_YEARS +
CHICAGO_HOME + QUAL_LEVEL + AFFINITY_SCORE + MG_PR_MODEL_DESC,
data = .)
Residuals:
Min 1Q Median 3Q Max
-4.2895 -0.6263 -0.1741 0.4893 5.4347
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.341446 0.111751 3.055 0.002251 **
ACTIVE_PROPOSALS -0.064470 0.032990 -1.954 0.050685 .
ns(NUMERIC_AGE, df = 5)1 0.356430 0.086139 4.138 3.52e-05 ***
ns(NUMERIC_AGE, df = 5)2 0.234530 0.100786 2.327 0.019975 *
ns(NUMERIC_AGE, df = 5)3 -0.367925 0.069120 -5.323 1.03e-07 ***
ns(NUMERIC_AGE, df = 5)4 0.434616 0.212833 2.042 0.041161 *
ns(NUMERIC_AGE, df = 5)5 -0.498620 0.113826 -4.381 1.19e-05 ***
PM_VISIT_LAST_2_YRS 0.311222 0.035286 8.820 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.274852 0.029943 9.179 < 2e-16 ***
AF_25K_GIFT 0.541953 0.043379 12.493 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.264019 0.006774 38.974 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.902994 0.042330 68.580 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.547836 0.025472 21.508 < 2e-16 ***
MG_250K_PLUS 0.839794 0.077224 10.875 < 2e-16 ***
PRESIDENT_VISIT 0.121050 0.062245 1.945 0.051820 .
TRUSTEE_OR_ADVISORY_BOARD -0.057299 0.032569 -1.759 0.078543 .
Alumnus -0.596510 0.024199 -24.650 < 2e-16 ***
DOUBLE_ALUM 0.011842 0.022958 0.516 0.605998
EVER_PARENT -0.056325 0.021948 -2.566 0.010288 *
SEASON_TICKET_YEARS -0.017930 0.004501 -3.984 6.81e-05 ***
CHICAGO_HOME -0.039806 0.016750 -2.376 0.017492 *
QUAL_LEVELA1 $100M+ 1.220711 0.484940 2.517 0.011836 *
QUAL_LEVELA2 $50M - 99.9M 1.434297 0.433305 3.310 0.000934 ***
QUAL_LEVELA3 $25M - $49.9M 0.417629 0.216237 1.931 0.053455 .
QUAL_LEVELA4 $10M - $24.9M 0.262164 0.149461 1.754 0.079436 .
QUAL_LEVELA5 $5M - $9.9M 0.462304 0.113449 4.075 4.62e-05 ***
QUAL_LEVELA6 $2M - $4.9M 0.388985 0.102558 3.793 0.000149 ***
QUAL_LEVELA7 $1M - $1.9M 0.286411 0.081213 3.527 0.000422 ***
QUAL_LEVELB $500K - $999K 0.065305 0.073322 0.891 0.373122
QUAL_LEVELC $250K - $499K -0.001726 0.069408 -0.025 0.980166
QUAL_LEVELD $100K - $249K -0.068684 0.068436 -1.004 0.315568
QUAL_LEVELE $50K - $99K -0.173363 0.070001 -2.477 0.013273 *
QUAL_LEVELF $25K - $49K 0.005171 0.070218 0.074 0.941294
QUAL_LEVELG $10K - $24K -0.159384 0.069455 -2.295 0.021756 *
QUAL_LEVELH Under $10K -0.281272 0.223546 -1.258 0.208324
QUAL_LEVELJ Future Prospect -1.277631 0.679280 -1.881 0.060006 .
AFFINITY_SCORE 0.169685 0.007186 23.612 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.116074 0.022853 -5.079 3.83e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.242416 0.028692 8.449 < 2e-16 ***
MG_PR_MODEL_DESCTop Tier 0.573753 0.029048 19.752 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9551 on 18738 degrees of freedom
Multiple R-squared: 0.7308, Adjusted R-squared: 0.7302
F-statistic: 1304 on 39 and 18738 DF, p-value: < 2.2e-16
[[2]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ACTIVE_PROPOSALS +
ns(NUMERIC_AGE, df = 5) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DOUBLE_ALUM + EVER_PARENT + SEASON_TICKET_YEARS +
CHICAGO_HOME + QUAL_LEVEL + AFFINITY_SCORE + MG_PR_MODEL_DESC,
data = .)
Residuals:
Min 1Q Median 3Q Max
-4.2897 -0.6220 -0.1744 0.4872 5.6408
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.391491 0.110456 3.544 0.000395 ***
ACTIVE_PROPOSALS -0.063937 0.033186 -1.927 0.054044 .
ns(NUMERIC_AGE, df = 5)1 0.328203 0.085189 3.853 0.000117 ***
ns(NUMERIC_AGE, df = 5)2 0.215162 0.099803 2.156 0.031107 *
ns(NUMERIC_AGE, df = 5)3 -0.387400 0.067545 -5.735 9.88e-09 ***
ns(NUMERIC_AGE, df = 5)4 0.416154 0.209315 1.988 0.046807 *
ns(NUMERIC_AGE, df = 5)5 -0.447092 0.103769 -4.309 1.65e-05 ***
PM_VISIT_LAST_2_YRS 0.305023 0.035400 8.616 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.273453 0.030004 9.114 < 2e-16 ***
AF_25K_GIFT 0.505943 0.044017 11.494 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.262934 0.006762 38.885 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.889784 0.042305 68.308 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.545837 0.025507 21.399 < 2e-16 ***
MG_250K_PLUS 0.846511 0.075645 11.191 < 2e-16 ***
PRESIDENT_VISIT 0.113156 0.062219 1.819 0.068979 .
TRUSTEE_OR_ADVISORY_BOARD -0.054670 0.032815 -1.666 0.095727 .
Alumnus -0.585837 0.024183 -24.225 < 2e-16 ***
DOUBLE_ALUM 0.014434 0.022814 0.633 0.526955
EVER_PARENT -0.047859 0.021898 -2.186 0.028862 *
SEASON_TICKET_YEARS -0.016981 0.004539 -3.741 0.000184 ***
CHICAGO_HOME -0.040877 0.016729 -2.443 0.014556 *
QUAL_LEVELA1 $100M+ 0.726674 0.433751 1.675 0.093887 .
QUAL_LEVELA2 $50M - 99.9M 1.122095 0.396099 2.833 0.004618 **
QUAL_LEVELA3 $25M - $49.9M 0.580742 0.216842 2.678 0.007409 **
QUAL_LEVELA4 $10M - $24.9M 0.367308 0.154011 2.385 0.017092 *
QUAL_LEVELA5 $5M - $9.9M 0.532593 0.111105 4.794 1.65e-06 ***
QUAL_LEVELA6 $2M - $4.9M 0.357393 0.102205 3.497 0.000472 ***
QUAL_LEVELA7 $1M - $1.9M 0.271406 0.080993 3.351 0.000807 ***
QUAL_LEVELB $500K - $999K 0.038988 0.072917 0.535 0.592873
QUAL_LEVELC $250K - $499K -0.033952 0.069026 -0.492 0.622817
QUAL_LEVELD $100K - $249K -0.102750 0.068018 -1.511 0.130900
QUAL_LEVELE $50K - $99K -0.206580 0.069616 -2.967 0.003007 **
QUAL_LEVELF $25K - $49K -0.016057 0.069841 -0.230 0.818167
QUAL_LEVELG $10K - $24K -0.174895 0.069020 -2.534 0.011286 *
QUAL_LEVELH Under $10K -0.280209 0.234367 -1.196 0.231868
QUAL_LEVELJ Future Prospect -1.308141 0.678928 -1.927 0.054023 .
AFFINITY_SCORE 0.169215 0.007183 23.557 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.121312 0.022845 -5.310 1.11e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.238070 0.028686 8.299 < 2e-16 ***
MG_PR_MODEL_DESCTop Tier 0.582833 0.029210 19.953 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9546 on 18738 degrees of freedom
Multiple R-squared: 0.7312, Adjusted R-squared: 0.7307
F-statistic: 1307 on 39 and 18738 DF, p-value: < 2.2e-16
[[3]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ACTIVE_PROPOSALS +
ns(NUMERIC_AGE, df = 5) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DOUBLE_ALUM + EVER_PARENT + SEASON_TICKET_YEARS +
CHICAGO_HOME + QUAL_LEVEL + AFFINITY_SCORE + MG_PR_MODEL_DESC,
data = .)
Residuals:
Min 1Q Median 3Q Max
-4.2725 -0.6257 -0.1754 0.4919 5.7364
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.404583 0.112542 3.595 0.000325 ***
ACTIVE_PROPOSALS -0.071400 0.033262 -2.147 0.031837 *
ns(NUMERIC_AGE, df = 5)1 0.285568 0.087031 3.281 0.001035 **
ns(NUMERIC_AGE, df = 5)2 0.151597 0.101923 1.487 0.136933
ns(NUMERIC_AGE, df = 5)3 -0.441464 0.069324 -6.368 1.96e-10 ***
ns(NUMERIC_AGE, df = 5)4 0.295610 0.215133 1.374 0.169434
ns(NUMERIC_AGE, df = 5)5 -0.458674 0.113424 -4.044 5.28e-05 ***
PM_VISIT_LAST_2_YRS 0.325636 0.035581 9.152 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.263200 0.029977 8.780 < 2e-16 ***
AF_25K_GIFT 0.536419 0.043587 12.307 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.266381 0.006785 39.260 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.895646 0.042392 68.307 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.553333 0.025609 21.607 < 2e-16 ***
MG_250K_PLUS 0.783007 0.076835 10.191 < 2e-16 ***
PRESIDENT_VISIT 0.113368 0.061956 1.830 0.067294 .
TRUSTEE_OR_ADVISORY_BOARD -0.056395 0.032916 -1.713 0.086672 .
Alumnus -0.593741 0.024216 -24.518 < 2e-16 ***
DOUBLE_ALUM 0.016607 0.022898 0.725 0.468308
EVER_PARENT -0.047143 0.021997 -2.143 0.032114 *
SEASON_TICKET_YEARS -0.018436 0.004462 -4.132 3.62e-05 ***
CHICAGO_HOME -0.040574 0.016835 -2.410 0.015959 *
QUAL_LEVELA1 $100M+ 0.768902 0.435343 1.766 0.077379 .
QUAL_LEVELA2 $50M - 99.9M 1.216988 0.484949 2.510 0.012098 *
QUAL_LEVELA3 $25M - $49.9M 0.661902 0.236010 2.805 0.005044 **
QUAL_LEVELA4 $10M - $24.9M 0.322639 0.150113 2.149 0.031623 *
QUAL_LEVELA5 $5M - $9.9M 0.504334 0.111437 4.526 6.06e-06 ***
QUAL_LEVELA6 $2M - $4.9M 0.424231 0.101644 4.174 3.01e-05 ***
QUAL_LEVELA7 $1M - $1.9M 0.307168 0.081771 3.756 0.000173 ***
QUAL_LEVELB $500K - $999K 0.055312 0.073250 0.755 0.450189
QUAL_LEVELC $250K - $499K 0.001848 0.069363 0.027 0.978740
QUAL_LEVELD $100K - $249K -0.065486 0.068360 -0.958 0.338096
QUAL_LEVELE $50K - $99K -0.174596 0.069988 -2.495 0.012616 *
QUAL_LEVELF $25K - $49K 0.008160 0.070161 0.116 0.907415
QUAL_LEVELG $10K - $24K -0.142123 0.069390 -2.048 0.040557 *
QUAL_LEVELH Under $10K -0.262234 0.224164 -1.170 0.242084
QUAL_LEVELJ Future Prospect -1.282014 0.681380 -1.881 0.059920 .
AFFINITY_SCORE 0.170403 0.007205 23.652 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.110027 0.022910 -4.803 1.58e-06 ***
MG_PR_MODEL_DESCMiddle Tier 0.233596 0.028819 8.105 5.57e-16 ***
MG_PR_MODEL_DESCTop Tier 0.573006 0.029232 19.602 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9581 on 18738 degrees of freedom
Multiple R-squared: 0.7297, Adjusted R-squared: 0.7291
F-statistic: 1297 on 39 and 18738 DF, p-value: < 2.2e-16
[[4]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ACTIVE_PROPOSALS +
ns(NUMERIC_AGE, df = 5) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DOUBLE_ALUM + EVER_PARENT + SEASON_TICKET_YEARS +
CHICAGO_HOME + QUAL_LEVEL + AFFINITY_SCORE + MG_PR_MODEL_DESC,
data = .)
Residuals:
Min 1Q Median 3Q Max
-4.2709 -0.6226 -0.1777 0.4889 5.6790
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.364453 0.112016 3.254 0.001142 **
ACTIVE_PROPOSALS -0.068631 0.033344 -2.058 0.039577 *
ns(NUMERIC_AGE, df = 5)1 0.342206 0.085931 3.982 6.85e-05 ***
ns(NUMERIC_AGE, df = 5)2 0.229831 0.100531 2.286 0.022255 *
ns(NUMERIC_AGE, df = 5)3 -0.425362 0.069194 -6.147 8.04e-10 ***
ns(NUMERIC_AGE, df = 5)4 0.476965 0.212390 2.246 0.024735 *
ns(NUMERIC_AGE, df = 5)5 -0.375766 0.114094 -3.293 0.000991 ***
PM_VISIT_LAST_2_YRS 0.316427 0.035724 8.857 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.285694 0.030000 9.523 < 2e-16 ***
AF_25K_GIFT 0.553407 0.043627 12.685 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.262587 0.006797 38.633 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.903585 0.042350 68.561 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.566545 0.025508 22.210 < 2e-16 ***
MG_250K_PLUS 0.782409 0.076957 10.167 < 2e-16 ***
PRESIDENT_VISIT 0.123001 0.062711 1.961 0.049848 *
TRUSTEE_OR_ADVISORY_BOARD -0.057048 0.033046 -1.726 0.084308 .
Alumnus -0.596085 0.024245 -24.586 < 2e-16 ***
DOUBLE_ALUM 0.004613 0.022957 0.201 0.840760
EVER_PARENT -0.063259 0.021905 -2.888 0.003883 **
SEASON_TICKET_YEARS -0.018392 0.004482 -4.103 4.09e-05 ***
CHICAGO_HOME -0.038783 0.016822 -2.306 0.021149 *
QUAL_LEVELA1 $100M+ 0.741595 0.435734 1.702 0.088782 .
QUAL_LEVELA2 $50M - 99.9M 1.155604 0.397917 2.904 0.003687 **
QUAL_LEVELA3 $25M - $49.9M 0.550423 0.217925 2.526 0.011553 *
QUAL_LEVELA4 $10M - $24.9M 0.395049 0.147932 2.670 0.007581 **
QUAL_LEVELA5 $5M - $9.9M 0.509263 0.111516 4.567 4.99e-06 ***
QUAL_LEVELA6 $2M - $4.9M 0.448581 0.104346 4.299 1.72e-05 ***
QUAL_LEVELA7 $1M - $1.9M 0.276262 0.081841 3.376 0.000738 ***
QUAL_LEVELB $500K - $999K 0.043400 0.073677 0.589 0.555826
QUAL_LEVELC $250K - $499K -0.002000 0.069700 -0.029 0.977108
QUAL_LEVELD $100K - $249K -0.065492 0.068717 -0.953 0.340565
QUAL_LEVELE $50K - $99K -0.175510 0.070339 -2.495 0.012597 *
QUAL_LEVELF $25K - $49K -0.002979 0.070550 -0.042 0.966316
QUAL_LEVELG $10K - $24K -0.155404 0.069806 -2.226 0.026012 *
QUAL_LEVELH Under $10K -0.292918 0.229810 -1.275 0.202465
QUAL_LEVELJ Future Prospect -1.604887 0.961722 -1.669 0.095181 .
AFFINITY_SCORE 0.168030 0.007220 23.272 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.120494 0.022974 -5.245 1.58e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.220163 0.028895 7.619 2.67e-14 ***
MG_PR_MODEL_DESCTop Tier 0.555617 0.029262 18.988 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9588 on 18738 degrees of freedom
Multiple R-squared: 0.7282, Adjusted R-squared: 0.7276
F-statistic: 1287 on 39 and 18738 DF, p-value: < 2.2e-16
[[5]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ACTIVE_PROPOSALS +
ns(NUMERIC_AGE, df = 5) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DOUBLE_ALUM + EVER_PARENT + SEASON_TICKET_YEARS +
CHICAGO_HOME + QUAL_LEVEL + AFFINITY_SCORE + MG_PR_MODEL_DESC,
data = .)
Residuals:
Min 1Q Median 3Q Max
-4.2347 -0.6301 -0.1769 0.4954 5.6709
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.385481 0.112055 3.440 0.000583 ***
ACTIVE_PROPOSALS -0.091010 0.033559 -2.712 0.006695 **
ns(NUMERIC_AGE, df = 5)1 0.315357 0.086665 3.639 0.000275 ***
ns(NUMERIC_AGE, df = 5)2 0.186084 0.101497 1.833 0.066760 .
ns(NUMERIC_AGE, df = 5)3 -0.405596 0.069642 -5.824 5.84e-09 ***
ns(NUMERIC_AGE, df = 5)4 0.369155 0.214564 1.720 0.085359 .
ns(NUMERIC_AGE, df = 5)5 -0.420228 0.115698 -3.632 0.000282 ***
PM_VISIT_LAST_2_YRS 0.351087 0.035817 9.802 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.276443 0.030245 9.140 < 2e-16 ***
AF_25K_GIFT 0.532387 0.043852 12.141 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.261329 0.006798 38.443 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.896785 0.042546 68.087 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.547849 0.025684 21.330 < 2e-16 ***
MG_250K_PLUS 0.800497 0.077118 10.380 < 2e-16 ***
PRESIDENT_VISIT 0.125092 0.062108 2.014 0.044012 *
TRUSTEE_OR_ADVISORY_BOARD -0.065094 0.032902 -1.978 0.047893 *
Alumnus -0.597361 0.024342 -24.541 < 2e-16 ***
DOUBLE_ALUM 0.016604 0.023189 0.716 0.473982
EVER_PARENT -0.057919 0.022132 -2.617 0.008879 **
SEASON_TICKET_YEARS -0.018980 0.004552 -4.170 3.06e-05 ***
CHICAGO_HOME -0.047823 0.016846 -2.839 0.004533 **
QUAL_LEVELA1 $100M+ 0.761863 0.437063 1.743 0.081325 .
QUAL_LEVELA2 $50M - 99.9M 1.170506 0.399146 2.933 0.003366 **
QUAL_LEVELA3 $25M - $49.9M 0.658111 0.232021 2.836 0.004567 **
QUAL_LEVELA4 $10M - $24.9M 0.354742 0.149205 2.378 0.017438 *
QUAL_LEVELA5 $5M - $9.9M 0.529395 0.111983 4.727 2.29e-06 ***
QUAL_LEVELA6 $2M - $4.9M 0.349239 0.101721 3.433 0.000598 ***
QUAL_LEVELA7 $1M - $1.9M 0.322173 0.082255 3.917 9.01e-05 ***
QUAL_LEVELB $500K - $999K 0.066680 0.074003 0.901 0.367577
QUAL_LEVELC $250K - $499K 0.010894 0.070094 0.155 0.876488
QUAL_LEVELD $100K - $249K -0.057196 0.069098 -0.828 0.407818
QUAL_LEVELE $50K - $99K -0.167359 0.070684 -2.368 0.017908 *
QUAL_LEVELF $25K - $49K 0.018070 0.070957 0.255 0.798988
QUAL_LEVELG $10K - $24K -0.150198 0.070064 -2.144 0.032067 *
QUAL_LEVELH Under $10K -0.278074 0.230630 -1.206 0.227942
QUAL_LEVELJ Future Prospect -1.281727 0.684037 -1.874 0.060978 .
AFFINITY_SCORE 0.170917 0.007221 23.669 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.115144 0.023018 -5.002 5.71e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.248133 0.028866 8.596 < 2e-16 ***
MG_PR_MODEL_DESCTop Tier 0.573315 0.029260 19.594 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9617 on 18738 degrees of freedom
Multiple R-squared: 0.727, Adjusted R-squared: 0.7264
F-statistic: 1279 on 39 and 18738 DF, p-value: < 2.2e-16
[[6]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ACTIVE_PROPOSALS +
ns(NUMERIC_AGE, df = 5) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DOUBLE_ALUM + EVER_PARENT + SEASON_TICKET_YEARS +
CHICAGO_HOME + QUAL_LEVEL + AFFINITY_SCORE + MG_PR_MODEL_DESC,
data = .)
Residuals:
Min 1Q Median 3Q Max
-4.2449 -0.6239 -0.1764 0.4922 5.6765
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.352804 0.111907 3.153 0.00162 **
ACTIVE_PROPOSALS -0.068250 0.033304 -2.049 0.04044 *
ns(NUMERIC_AGE, df = 5)1 0.373898 0.085409 4.378 1.21e-05 ***
ns(NUMERIC_AGE, df = 5)2 0.214894 0.100002 2.149 0.03166 *
ns(NUMERIC_AGE, df = 5)3 -0.388711 0.069060 -5.629 1.84e-08 ***
ns(NUMERIC_AGE, df = 5)4 0.440838 0.211422 2.085 0.03707 *
ns(NUMERIC_AGE, df = 5)5 -0.455911 0.114509 -3.981 6.88e-05 ***
PM_VISIT_LAST_2_YRS 0.283859 0.035678 7.956 1.87e-15 ***
log10plus1(VISIT_COUNT) 0.303688 0.029989 10.127 < 2e-16 ***
AF_25K_GIFT 0.544760 0.044284 12.301 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.263631 0.006790 38.828 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.889769 0.042429 68.108 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.551496 0.025512 21.617 < 2e-16 ***
MG_250K_PLUS 0.841226 0.077876 10.802 < 2e-16 ***
PRESIDENT_VISIT 0.107264 0.062923 1.705 0.08827 .
TRUSTEE_OR_ADVISORY_BOARD -0.070653 0.032772 -2.156 0.03111 *
Alumnus -0.600663 0.024199 -24.822 < 2e-16 ***
DOUBLE_ALUM 0.004623 0.022959 0.201 0.84044
EVER_PARENT -0.054865 0.021972 -2.497 0.01253 *
SEASON_TICKET_YEARS -0.018012 0.004483 -4.018 5.90e-05 ***
CHICAGO_HOME -0.034621 0.016758 -2.066 0.03885 *
QUAL_LEVELA1 $100M+ 0.708448 0.484251 1.463 0.14349
QUAL_LEVELA2 $50M - 99.9M 1.159842 0.397173 2.920 0.00350 **
QUAL_LEVELA3 $25M - $49.9M 0.559074 0.213457 2.619 0.00882 **
QUAL_LEVELA4 $10M - $24.9M 0.371337 0.152003 2.443 0.01458 *
QUAL_LEVELA5 $5M - $9.9M 0.553517 0.113532 4.875 1.09e-06 ***
QUAL_LEVELA6 $2M - $4.9M 0.330123 0.104017 3.174 0.00151 **
QUAL_LEVELA7 $1M - $1.9M 0.327043 0.082665 3.956 7.64e-05 ***
QUAL_LEVELB $500K - $999K 0.057462 0.074712 0.769 0.44183
QUAL_LEVELC $250K - $499K 0.007468 0.070772 0.106 0.91596
QUAL_LEVELD $100K - $249K -0.050540 0.069833 -0.724 0.46924
QUAL_LEVELE $50K - $99K -0.164136 0.071409 -2.299 0.02154 *
QUAL_LEVELF $25K - $49K 0.011728 0.071635 0.164 0.86996
QUAL_LEVELG $10K - $24K -0.140502 0.070838 -1.983 0.04733 *
QUAL_LEVELH Under $10K -0.531455 0.235345 -2.258 0.02394 *
QUAL_LEVELJ Future Prospect -1.264960 0.680433 -1.859 0.06304 .
AFFINITY_SCORE 0.170109 0.007208 23.601 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.131847 0.022948 -5.745 9.31e-09 ***
MG_PR_MODEL_DESCMiddle Tier 0.217021 0.028800 7.535 5.09e-14 ***
MG_PR_MODEL_DESCTop Tier 0.543232 0.029141 18.641 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9565 on 18738 degrees of freedom
Multiple R-squared: 0.7291, Adjusted R-squared: 0.7286
F-statistic: 1293 on 39 and 18738 DF, p-value: < 2.2e-16
[[7]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ACTIVE_PROPOSALS +
ns(NUMERIC_AGE, df = 5) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DOUBLE_ALUM + EVER_PARENT + SEASON_TICKET_YEARS +
CHICAGO_HOME + QUAL_LEVEL + AFFINITY_SCORE + MG_PR_MODEL_DESC,
data = .)
Residuals:
Min 1Q Median 3Q Max
-4.2851 -0.6251 -0.1747 0.4941 5.6212
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.346854 0.110443 3.141 0.001689 **
ACTIVE_PROPOSALS -0.105155 0.033343 -3.154 0.001614 **
ns(NUMERIC_AGE, df = 5)1 0.320110 0.085101 3.762 0.000169 ***
ns(NUMERIC_AGE, df = 5)2 0.200040 0.099742 2.006 0.044917 *
ns(NUMERIC_AGE, df = 5)3 -0.464293 0.069248 -6.705 2.07e-11 ***
ns(NUMERIC_AGE, df = 5)4 0.427210 0.210857 2.026 0.042773 *
ns(NUMERIC_AGE, df = 5)5 -0.364465 0.115968 -3.143 0.001676 **
PM_VISIT_LAST_2_YRS 0.329574 0.035547 9.271 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.301811 0.030020 10.054 < 2e-16 ***
AF_25K_GIFT 0.552387 0.043573 12.677 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.266517 0.006781 39.305 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.878919 0.042437 67.839 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.557458 0.025566 21.804 < 2e-16 ***
MG_250K_PLUS 0.796596 0.077577 10.268 < 2e-16 ***
PRESIDENT_VISIT 0.102970 0.062228 1.655 0.097997 .
TRUSTEE_OR_ADVISORY_BOARD -0.051510 0.032666 -1.577 0.114844
Alumnus -0.589388 0.024322 -24.233 < 2e-16 ***
DOUBLE_ALUM 0.025664 0.023129 1.110 0.267177
EVER_PARENT -0.055094 0.022095 -2.493 0.012659 *
SEASON_TICKET_YEARS -0.017291 0.004571 -3.783 0.000156 ***
CHICAGO_HOME -0.036867 0.016776 -2.198 0.027986 *
QUAL_LEVELA1 $100M+ 0.792449 0.435250 1.821 0.068672 .
QUAL_LEVELA2 $50M - 99.9M 0.920865 0.484359 1.901 0.057291 .
QUAL_LEVELA3 $25M - $49.9M 0.586341 0.225934 2.595 0.009461 **
QUAL_LEVELA4 $10M - $24.9M 0.372010 0.150190 2.477 0.013261 *
QUAL_LEVELA5 $5M - $9.9M 0.583059 0.110893 5.258 1.47e-07 ***
QUAL_LEVELA6 $2M - $4.9M 0.356838 0.102873 3.469 0.000524 ***
QUAL_LEVELA7 $1M - $1.9M 0.322053 0.081231 3.965 7.38e-05 ***
QUAL_LEVELB $500K - $999K 0.069831 0.073386 0.952 0.341330
QUAL_LEVELC $250K - $499K 0.016917 0.069385 0.244 0.807375
QUAL_LEVELD $100K - $249K -0.051149 0.068407 -0.748 0.454635
QUAL_LEVELE $50K - $99K -0.156419 0.070005 -2.234 0.025468 *
QUAL_LEVELF $25K - $49K 0.015232 0.070271 0.217 0.828401
QUAL_LEVELG $10K - $24K -0.132805 0.069396 -1.914 0.055670 .
QUAL_LEVELH Under $10K -0.203682 0.241516 -0.843 0.399045
QUAL_LEVELJ Future Prospect -0.944384 0.961149 -0.983 0.325838
AFFINITY_SCORE 0.168162 0.007203 23.346 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.107480 0.022946 -4.684 2.83e-06 ***
MG_PR_MODEL_DESCMiddle Tier 0.238380 0.028842 8.265 < 2e-16 ***
MG_PR_MODEL_DESCTop Tier 0.557588 0.029195 19.099 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9578 on 18738 degrees of freedom
Multiple R-squared: 0.7284, Adjusted R-squared: 0.7279
F-statistic: 1289 on 39 and 18738 DF, p-value: < 2.2e-16
[[8]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ACTIVE_PROPOSALS +
ns(NUMERIC_AGE, df = 5) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DOUBLE_ALUM + EVER_PARENT + SEASON_TICKET_YEARS +
CHICAGO_HOME + QUAL_LEVEL + AFFINITY_SCORE + MG_PR_MODEL_DESC,
data = .)
Residuals:
Min 1Q Median 3Q Max
-4.2807 -0.6268 -0.1746 0.4972 5.6138
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.382629 0.113022 3.385 0.000712 ***
ACTIVE_PROPOSALS -0.063374 0.033423 -1.896 0.057960 .
ns(NUMERIC_AGE, df = 5)1 0.377380 0.086519 4.362 1.30e-05 ***
ns(NUMERIC_AGE, df = 5)2 0.274991 0.101330 2.714 0.006658 **
ns(NUMERIC_AGE, df = 5)3 -0.403471 0.069412 -5.813 6.25e-09 ***
ns(NUMERIC_AGE, df = 5)4 0.536897 0.213888 2.510 0.012075 *
ns(NUMERIC_AGE, df = 5)5 -0.361208 0.113673 -3.178 0.001487 **
PM_VISIT_LAST_2_YRS 0.314056 0.035897 8.749 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.286774 0.030229 9.487 < 2e-16 ***
AF_25K_GIFT 0.508165 0.044300 11.471 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.261251 0.006836 38.218 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.876721 0.042474 67.729 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.544274 0.025720 21.162 < 2e-16 ***
MG_250K_PLUS 0.818358 0.077474 10.563 < 2e-16 ***
PRESIDENT_VISIT 0.105752 0.063399 1.668 0.095328 .
TRUSTEE_OR_ADVISORY_BOARD -0.067086 0.033334 -2.013 0.044176 *
Alumnus -0.596046 0.024415 -24.413 < 2e-16 ***
DOUBLE_ALUM 0.027571 0.023032 1.197 0.231309
EVER_PARENT -0.064839 0.022138 -2.929 0.003406 **
SEASON_TICKET_YEARS -0.017011 0.004525 -3.760 0.000171 ***
CHICAGO_HOME -0.048085 0.016896 -2.846 0.004432 **
QUAL_LEVELA1 $100M+ 0.209403 0.486964 0.430 0.667188
QUAL_LEVELA2 $50M - 99.9M 1.092914 0.399484 2.736 0.006228 **
QUAL_LEVELA3 $25M - $49.9M 0.559822 0.237247 2.360 0.018302 *
QUAL_LEVELA4 $10M - $24.9M 0.334559 0.150036 2.230 0.025769 *
QUAL_LEVELA5 $5M - $9.9M 0.495923 0.115704 4.286 1.83e-05 ***
QUAL_LEVELA6 $2M - $4.9M 0.346749 0.104109 3.331 0.000868 ***
QUAL_LEVELA7 $1M - $1.9M 0.226542 0.083347 2.718 0.006572 **
QUAL_LEVELB $500K - $999K -0.011676 0.075139 -0.155 0.876516
QUAL_LEVELC $250K - $499K -0.063212 0.071198 -0.888 0.374640
QUAL_LEVELD $100K - $249K -0.126086 0.070233 -1.795 0.072629 .
QUAL_LEVELE $50K - $99K -0.242572 0.071821 -3.377 0.000733 ***
QUAL_LEVELF $25K - $49K -0.059704 0.072044 -0.829 0.407273
QUAL_LEVELG $10K - $24K -0.210236 0.071221 -2.952 0.003162 **
QUAL_LEVELH Under $10K -0.197083 0.243128 -0.811 0.417596
QUAL_LEVELJ Future Prospect -1.347452 0.684371 -1.969 0.048980 *
AFFINITY_SCORE 0.171930 0.007249 23.719 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.110816 0.022981 -4.822 1.43e-06 ***
MG_PR_MODEL_DESCMiddle Tier 0.230762 0.028902 7.984 1.49e-15 ***
MG_PR_MODEL_DESCTop Tier 0.571219 0.029310 19.489 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.962 on 18738 degrees of freedom
Multiple R-squared: 0.7257, Adjusted R-squared: 0.7251
F-statistic: 1271 on 39 and 18738 DF, p-value: < 2.2e-16
[[9]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ACTIVE_PROPOSALS +
ns(NUMERIC_AGE, df = 5) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DOUBLE_ALUM + EVER_PARENT + SEASON_TICKET_YEARS +
CHICAGO_HOME + QUAL_LEVEL + AFFINITY_SCORE + MG_PR_MODEL_DESC,
data = .)
Residuals:
Min 1Q Median 3Q Max
-4.2872 -0.6215 -0.1758 0.4889 5.6850
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.346765 0.111376 3.113 0.001852 **
ACTIVE_PROPOSALS -0.063054 0.032974 -1.912 0.055861 .
ns(NUMERIC_AGE, df = 5)1 0.366713 0.085852 4.271 1.95e-05 ***
ns(NUMERIC_AGE, df = 5)2 0.250321 0.100573 2.489 0.012821 *
ns(NUMERIC_AGE, df = 5)3 -0.379278 0.069240 -5.478 4.36e-08 ***
ns(NUMERIC_AGE, df = 5)4 0.483587 0.212254 2.278 0.022717 *
ns(NUMERIC_AGE, df = 5)5 -0.417211 0.114763 -3.635 0.000278 ***
PM_VISIT_LAST_2_YRS 0.321170 0.035279 9.104 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.274981 0.030052 9.150 < 2e-16 ***
AF_25K_GIFT 0.499074 0.043796 11.396 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.263008 0.006806 38.641 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.891665 0.042624 67.842 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.564097 0.025582 22.050 < 2e-16 ***
MG_250K_PLUS 0.770537 0.076851 10.026 < 2e-16 ***
PRESIDENT_VISIT 0.135153 0.062022 2.179 0.029337 *
TRUSTEE_OR_ADVISORY_BOARD -0.074513 0.032861 -2.268 0.023370 *
Alumnus -0.595276 0.024279 -24.518 < 2e-16 ***
DOUBLE_ALUM 0.012056 0.023024 0.524 0.600535
EVER_PARENT -0.059858 0.022081 -2.711 0.006718 **
SEASON_TICKET_YEARS -0.020870 0.004500 -4.637 3.55e-06 ***
CHICAGO_HOME -0.042632 0.016811 -2.536 0.011225 *
QUAL_LEVELA1 $100M+ 0.809436 0.559844 1.446 0.148243
QUAL_LEVELA2 $50M - 99.9M 1.145850 0.398099 2.878 0.004003 **
QUAL_LEVELA3 $25M - $49.9M 0.509717 0.220884 2.308 0.021031 *
QUAL_LEVELA4 $10M - $24.9M 0.287625 0.148298 1.940 0.052454 .
QUAL_LEVELA5 $5M - $9.9M 0.507927 0.111796 4.543 5.57e-06 ***
QUAL_LEVELA6 $2M - $4.9M 0.384498 0.102731 3.743 0.000183 ***
QUAL_LEVELA7 $1M - $1.9M 0.278720 0.082053 3.397 0.000683 ***
QUAL_LEVELB $500K - $999K 0.036884 0.073741 0.500 0.616947
QUAL_LEVELC $250K - $499K -0.031306 0.069941 -0.448 0.654445
QUAL_LEVELD $100K - $249K -0.080208 0.068895 -1.164 0.244359
QUAL_LEVELE $50K - $99K -0.198185 0.070493 -2.811 0.004938 **
QUAL_LEVELF $25K - $49K -0.011455 0.070748 -0.162 0.871378
QUAL_LEVELG $10K - $24K -0.173836 0.069913 -2.486 0.012911 *
QUAL_LEVELH Under $10K -0.302310 0.219654 -1.376 0.168745
QUAL_LEVELJ Future Prospect -1.301232 0.682217 -1.907 0.056489 .
AFFINITY_SCORE 0.171624 0.007207 23.813 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.116209 0.023034 -5.045 4.57e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.236247 0.028984 8.151 3.84e-16 ***
MG_PR_MODEL_DESCTop Tier 0.569562 0.029192 19.511 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9592 on 18738 degrees of freedom
Multiple R-squared: 0.7293, Adjusted R-squared: 0.7288
F-statistic: 1295 on 39 and 18738 DF, p-value: < 2.2e-16
[[10]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ACTIVE_PROPOSALS +
ns(NUMERIC_AGE, df = 5) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DOUBLE_ALUM + EVER_PARENT + SEASON_TICKET_YEARS +
CHICAGO_HOME + QUAL_LEVEL + AFFINITY_SCORE + MG_PR_MODEL_DESC,
data = .)
Residuals:
Min 1Q Median 3Q Max
-4.2873 -0.6301 -0.1801 0.4971 5.7257
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.3122695 0.1110545 2.812 0.004931 **
ACTIVE_PROPOSALS -0.0736031 0.0333534 -2.207 0.027343 *
ns(NUMERIC_AGE, df = 5)1 0.3488614 0.0855520 4.078 4.57e-05 ***
ns(NUMERIC_AGE, df = 5)2 0.2494582 0.1001716 2.490 0.012772 *
ns(NUMERIC_AGE, df = 5)3 -0.3885250 0.0691745 -5.617 1.98e-08 ***
ns(NUMERIC_AGE, df = 5)4 0.4906668 0.2116286 2.319 0.020431 *
ns(NUMERIC_AGE, df = 5)5 -0.4295347 0.1138515 -3.773 0.000162 ***
PM_VISIT_LAST_2_YRS 0.3153022 0.0355707 8.864 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.2877440 0.0301351 9.548 < 2e-16 ***
AF_25K_GIFT 0.5417772 0.0437565 12.382 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.2578015 0.0068164 37.821 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.8997166 0.0426106 68.052 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.5710922 0.0256487 22.266 < 2e-16 ***
MG_250K_PLUS 0.7969775 0.0769718 10.354 < 2e-16 ***
PRESIDENT_VISIT 0.0763249 0.0629863 1.212 0.225616
TRUSTEE_OR_ADVISORY_BOARD -0.0534279 0.0329842 -1.620 0.105292
Alumnus -0.5849263 0.0243562 -24.015 < 2e-16 ***
DOUBLE_ALUM -0.0009347 0.0230807 -0.040 0.967699
EVER_PARENT -0.0542565 0.0221380 -2.451 0.014262 *
SEASON_TICKET_YEARS -0.0206457 0.0045570 -4.531 5.92e-06 ***
CHICAGO_HOME -0.0309689 0.0168936 -1.833 0.066792 .
QUAL_LEVELA1 $100M+ 0.7871819 0.4372446 1.800 0.071826 .
QUAL_LEVELA2 $50M - 99.9M 1.0705242 0.4359048 2.456 0.014064 *
QUAL_LEVELA3 $25M - $49.9M 0.5868876 0.2214272 2.650 0.008045 **
QUAL_LEVELA4 $10M - $24.9M 0.3578239 0.1493322 2.396 0.016578 *
QUAL_LEVELA5 $5M - $9.9M 0.4941051 0.1127425 4.383 1.18e-05 ***
QUAL_LEVELA6 $2M - $4.9M 0.3691003 0.1018378 3.624 0.000290 ***
QUAL_LEVELA7 $1M - $1.9M 0.2948647 0.0820841 3.592 0.000329 ***
QUAL_LEVELB $500K - $999K 0.0633141 0.0736615 0.860 0.390060
QUAL_LEVELC $250K - $499K 0.0213256 0.0698541 0.305 0.760151
QUAL_LEVELD $100K - $249K -0.0457723 0.0687890 -0.665 0.505802
QUAL_LEVELE $50K - $99K -0.1586832 0.0704319 -2.253 0.024271 *
QUAL_LEVELF $25K - $49K 0.0178503 0.0706532 0.253 0.800544
QUAL_LEVELG $10K - $24K -0.1331537 0.0698108 -1.907 0.056491 .
QUAL_LEVELH Under $10K -0.2728308 0.2252079 -1.211 0.225734
QUAL_LEVELJ Future Prospect -1.2680291 0.6843323 -1.853 0.063906 .
AFFINITY_SCORE 0.1727521 0.0072285 23.899 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.1110314 0.0231323 -4.800 1.60e-06 ***
MG_PR_MODEL_DESCMiddle Tier 0.2290186 0.0290201 7.892 3.14e-15 ***
MG_PR_MODEL_DESCTop Tier 0.5568048 0.0293331 18.982 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9622 on 18734 degrees of freedom
Multiple R-squared: 0.7265, Adjusted R-squared: 0.7259
F-statistic: 1276 on 39 and 18734 DF, p-value: < 2.2e-16Campaign all predictors 2
lapply(clmaps2, function(x) summary(x))[[1]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = 5) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4291 -0.6237 -0.1842 0.4958 5.4709
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.255115 0.094156 2.709 0.00675 **
ns(NUMERIC_AGE, df = 5)1 0.386056 0.086031 4.487 7.25e-06 ***
ns(NUMERIC_AGE, df = 5)2 0.250456 0.100715 2.487 0.01290 *
ns(NUMERIC_AGE, df = 5)3 -0.347147 0.069289 -5.010 5.49e-07 ***
ns(NUMERIC_AGE, df = 5)4 0.464897 0.213109 2.181 0.02916 *
ns(NUMERIC_AGE, df = 5)5 -0.546338 0.114141 -4.787 1.71e-06 ***
PM_VISIT_LAST_2_YRS 0.311628 0.028735 10.845 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.274347 0.028878 9.500 < 2e-16 ***
AF_25K_GIFT 0.615943 0.042566 14.470 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.260803 0.006777 38.485 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.929400 0.042294 69.262 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.558071 0.025498 21.887 < 2e-16 ***
MG_250K_PLUS 1.125296 0.071607 15.715 < 2e-16 ***
Alumnus -0.572103 0.021390 -26.746 < 2e-16 ***
SEASON_TICKET_YEARS -0.019221 0.004461 -4.309 1.65e-05 ***
AFFINITY_SCORE 0.158247 0.006798 23.279 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.117862 0.022128 -5.326 1.01e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.214713 0.028350 7.574 3.80e-14 ***
MG_PR_MODEL_DESCTop Tier 0.595287 0.028195 21.113 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9611 on 18759 degrees of freedom
Multiple R-squared: 0.7271, Adjusted R-squared: 0.7268
F-statistic: 2776 on 18 and 18759 DF, p-value: < 2.2e-16
[[2]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = 5) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4727 -0.6214 -0.1825 0.4944 6.2743
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.283317 0.093030 3.045 0.00233 **
ns(NUMERIC_AGE, df = 5)1 0.359258 0.085039 4.225 2.40e-05 ***
ns(NUMERIC_AGE, df = 5)2 0.230563 0.099705 2.312 0.02076 *
ns(NUMERIC_AGE, df = 5)3 -0.369434 0.067707 -5.456 4.92e-08 ***
ns(NUMERIC_AGE, df = 5)4 0.450814 0.209545 2.151 0.03146 *
ns(NUMERIC_AGE, df = 5)5 -0.482863 0.104079 -4.639 3.52e-06 ***
PM_VISIT_LAST_2_YRS 0.302190 0.028790 10.496 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.279258 0.028960 9.643 < 2e-16 ***
AF_25K_GIFT 0.588082 0.043198 13.614 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.259590 0.006763 38.383 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.913908 0.042293 68.898 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.557706 0.025538 21.839 < 2e-16 ***
MG_250K_PLUS 1.166991 0.070045 16.661 < 2e-16 ***
Alumnus -0.568739 0.021427 -26.543 < 2e-16 ***
SEASON_TICKET_YEARS -0.018724 0.004505 -4.156 3.25e-05 ***
AFFINITY_SCORE 0.158592 0.006794 23.343 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.123273 0.022106 -5.576 2.49e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.208271 0.028348 7.347 2.11e-13 ***
MG_PR_MODEL_DESCTop Tier 0.600171 0.028343 21.175 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.961 on 18759 degrees of freedom
Multiple R-squared: 0.7273, Adjusted R-squared: 0.7271
F-statistic: 2780 on 18 and 18759 DF, p-value: < 2.2e-16
[[3]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = 5) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3972 -0.6247 -0.1861 0.4948 6.3062
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.331319 0.095104 3.484 0.000496 ***
ns(NUMERIC_AGE, df = 5)1 0.310424 0.086897 3.572 0.000355 ***
ns(NUMERIC_AGE, df = 5)2 0.161271 0.101850 1.583 0.113343
ns(NUMERIC_AGE, df = 5)3 -0.422841 0.069459 -6.088 1.17e-09 ***
ns(NUMERIC_AGE, df = 5)4 0.320024 0.215424 1.486 0.137415
ns(NUMERIC_AGE, df = 5)5 -0.497399 0.113736 -4.373 1.23e-05 ***
PM_VISIT_LAST_2_YRS 0.319925 0.028793 11.111 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.263066 0.028893 9.105 < 2e-16 ***
AF_25K_GIFT 0.613941 0.042757 14.359 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.262874 0.006788 38.726 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.916592 0.042389 68.805 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.563381 0.025631 21.980 < 2e-16 ***
MG_250K_PLUS 1.089595 0.071116 15.321 < 2e-16 ***
Alumnus -0.574934 0.021439 -26.818 < 2e-16 ***
SEASON_TICKET_YEARS -0.020092 0.004420 -4.546 5.50e-06 ***
AFFINITY_SCORE 0.160383 0.006823 23.506 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.110457 0.022177 -4.981 6.39e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.207385 0.028463 7.286 3.32e-13 ***
MG_PR_MODEL_DESCTop Tier 0.594950 0.028351 20.985 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.964 on 18759 degrees of freedom
Multiple R-squared: 0.726, Adjusted R-squared: 0.7257
F-statistic: 2761 on 18 and 18759 DF, p-value: < 2.2e-16
[[4]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = 5) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4015 -0.6252 -0.1848 0.4979 6.2642
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.274694 0.093894 2.926 0.003442 **
ns(NUMERIC_AGE, df = 5)1 0.375171 0.085767 4.374 1.22e-05 ***
ns(NUMERIC_AGE, df = 5)2 0.250632 0.100418 2.496 0.012573 *
ns(NUMERIC_AGE, df = 5)3 -0.405832 0.069359 -5.851 4.96e-09 ***
ns(NUMERIC_AGE, df = 5)4 0.526632 0.212578 2.477 0.013245 *
ns(NUMERIC_AGE, df = 5)5 -0.418507 0.114422 -3.658 0.000255 ***
PM_VISIT_LAST_2_YRS 0.319594 0.028956 11.037 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.287850 0.028926 9.951 < 2e-16 ***
AF_25K_GIFT 0.625386 0.042895 14.579 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.259398 0.006796 38.169 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.925863 0.042344 69.098 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.578451 0.025528 22.659 < 2e-16 ***
MG_250K_PLUS 1.092788 0.071386 15.308 < 2e-16 ***
Alumnus -0.572718 0.021470 -26.675 < 2e-16 ***
SEASON_TICKET_YEARS -0.019652 0.004442 -4.424 9.76e-06 ***
AFFINITY_SCORE 0.155681 0.006829 22.797 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.124106 0.022225 -5.584 2.38e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.192985 0.028538 6.762 1.40e-11 ***
MG_PR_MODEL_DESCTop Tier 0.576692 0.028387 20.315 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9648 on 18759 degrees of freedom
Multiple R-squared: 0.7244, Adjusted R-squared: 0.7242
F-statistic: 2740 on 18 and 18759 DF, p-value: < 2.2e-16
[[5]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = 5) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.1069 -0.6275 -0.1859 0.4998 6.2731
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.304096 0.094649 3.213 0.00132 **
ns(NUMERIC_AGE, df = 5)1 0.349770 0.086479 4.045 5.26e-05 ***
ns(NUMERIC_AGE, df = 5)2 0.205294 0.101354 2.026 0.04283 *
ns(NUMERIC_AGE, df = 5)3 -0.385610 0.069824 -5.523 3.38e-08 ***
ns(NUMERIC_AGE, df = 5)4 0.411601 0.214714 1.917 0.05526 .
ns(NUMERIC_AGE, df = 5)5 -0.467467 0.116063 -4.028 5.65e-05 ***
PM_VISIT_LAST_2_YRS 0.339281 0.029088 11.664 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.276994 0.029135 9.507 < 2e-16 ***
AF_25K_GIFT 0.607389 0.043052 14.108 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.258651 0.006802 38.026 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.917820 0.042572 68.539 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.559188 0.025716 21.745 < 2e-16 ***
MG_250K_PLUS 1.097125 0.071871 15.265 < 2e-16 ***
Alumnus -0.573888 0.021552 -26.628 < 2e-16 ***
SEASON_TICKET_YEARS -0.020764 0.004515 -4.599 4.26e-06 ***
AFFINITY_SCORE 0.158616 0.006845 23.171 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.116487 0.022304 -5.223 1.78e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.221006 0.028528 7.747 9.88e-15 ***
MG_PR_MODEL_DESCTop Tier 0.594491 0.028418 20.920 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9682 on 18759 degrees of freedom
Multiple R-squared: 0.723, Adjusted R-squared: 0.7227
F-statistic: 2720 on 18 and 18759 DF, p-value: < 2.2e-16
[[6]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = 5) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4361 -0.6200 -0.1867 0.4945 6.2946
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.278270 0.093230 2.985 0.00284 **
ns(NUMERIC_AGE, df = 5)1 0.410210 0.085208 4.814 1.49e-06 ***
ns(NUMERIC_AGE, df = 5)2 0.233923 0.099899 2.342 0.01921 *
ns(NUMERIC_AGE, df = 5)3 -0.361744 0.069187 -5.229 1.73e-07 ***
ns(NUMERIC_AGE, df = 5)4 0.484563 0.211617 2.290 0.02204 *
ns(NUMERIC_AGE, df = 5)5 -0.497700 0.114815 -4.335 1.47e-05 ***
PM_VISIT_LAST_2_YRS 0.283623 0.029129 9.737 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.300906 0.028885 10.417 < 2e-16 ***
AF_25K_GIFT 0.614608 0.043478 14.136 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.260185 0.006788 38.331 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.913073 0.042421 68.670 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.562183 0.025537 22.014 < 2e-16 ***
MG_250K_PLUS 1.139511 0.072230 15.776 < 2e-16 ***
Alumnus -0.581322 0.021355 -27.222 < 2e-16 ***
SEASON_TICKET_YEARS -0.018984 0.004442 -4.274 1.93e-05 ***
AFFINITY_SCORE 0.158698 0.006821 23.265 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.132689 0.022198 -5.978 2.31e-09 ***
MG_PR_MODEL_DESCMiddle Tier 0.191770 0.028444 6.742 1.61e-11 ***
MG_PR_MODEL_DESCTop Tier 0.565934 0.028285 20.008 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9624 on 18759 degrees of freedom
Multiple R-squared: 0.7255, Adjusted R-squared: 0.7252
F-statistic: 2754 on 18 and 18759 DF, p-value: < 2.2e-16
[[7]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = 5) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3994 -0.6266 -0.1874 0.4958 6.2636
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.283086 0.093027 3.043 0.002345 **
ns(NUMERIC_AGE, df = 5)1 0.349749 0.084895 4.120 3.81e-05 ***
ns(NUMERIC_AGE, df = 5)2 0.214313 0.099606 2.152 0.031441 *
ns(NUMERIC_AGE, df = 5)3 -0.438830 0.069369 -6.326 2.57e-10 ***
ns(NUMERIC_AGE, df = 5)4 0.456633 0.211028 2.164 0.030489 *
ns(NUMERIC_AGE, df = 5)5 -0.411888 0.116315 -3.541 0.000399 ***
PM_VISIT_LAST_2_YRS 0.305546 0.028936 10.560 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.299356 0.028897 10.359 < 2e-16 ***
AF_25K_GIFT 0.623798 0.042930 14.531 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.263053 0.006784 38.776 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.899286 0.042408 68.366 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.567406 0.025584 22.179 < 2e-16 ***
MG_250K_PLUS 1.092580 0.072594 15.051 < 2e-16 ***
Alumnus -0.570026 0.021460 -26.562 < 2e-16 ***
SEASON_TICKET_YEARS -0.018948 0.004532 -4.181 2.92e-05 ***
AFFINITY_SCORE 0.157965 0.006824 23.149 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.109675 0.022215 -4.937 8.01e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.212359 0.028477 7.457 9.23e-14 ***
MG_PR_MODEL_DESCTop Tier 0.578368 0.028332 20.414 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9637 on 18759 degrees of freedom
Multiple R-squared: 0.7248, Adjusted R-squared: 0.7245
F-statistic: 2745 on 18 and 18759 DF, p-value: < 2.2e-16
[[8]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = 5) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4493 -0.6252 -0.1804 0.4957 6.2555
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.242819 0.094508 2.569 0.010198 *
ns(NUMERIC_AGE, df = 5)1 0.400320 0.086354 4.636 3.58e-06 ***
ns(NUMERIC_AGE, df = 5)2 0.280377 0.101216 2.770 0.005610 **
ns(NUMERIC_AGE, df = 5)3 -0.386612 0.069558 -5.558 2.76e-08 ***
ns(NUMERIC_AGE, df = 5)4 0.554027 0.214110 2.588 0.009673 **
ns(NUMERIC_AGE, df = 5)5 -0.417324 0.114021 -3.660 0.000253 ***
PM_VISIT_LAST_2_YRS 0.318603 0.029185 10.917 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.287104 0.029101 9.866 < 2e-16 ***
AF_25K_GIFT 0.576878 0.043685 13.206 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.258682 0.006836 37.844 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.901273 0.042488 68.284 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.555337 0.025751 21.565 < 2e-16 ***
MG_250K_PLUS 1.127600 0.072302 15.596 < 2e-16 ***
Alumnus -0.568027 0.021564 -26.341 < 2e-16 ***
SEASON_TICKET_YEARS -0.018766 0.004485 -4.185 2.87e-05 ***
AFFINITY_SCORE 0.160057 0.006853 23.356 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.117462 0.022260 -5.277 1.33e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.200356 0.028555 7.016 2.35e-12 ***
MG_PR_MODEL_DESCTop Tier 0.585297 0.028422 20.593 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9682 on 18759 degrees of freedom
Multiple R-squared: 0.7218, Adjusted R-squared: 0.7216
F-statistic: 2704 on 18 and 18759 DF, p-value: < 2.2e-16
[[9]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = 5) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3846 -0.6212 -0.1874 0.4930 6.2869
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.240996 0.093787 2.570 0.0102 *
ns(NUMERIC_AGE, df = 5)1 0.398169 0.085670 4.648 3.38e-06 ***
ns(NUMERIC_AGE, df = 5)2 0.265551 0.100441 2.644 0.0082 **
ns(NUMERIC_AGE, df = 5)3 -0.358989 0.069381 -5.174 2.31e-07 ***
ns(NUMERIC_AGE, df = 5)4 0.521546 0.212440 2.455 0.0141 *
ns(NUMERIC_AGE, df = 5)5 -0.457178 0.115156 -3.970 7.21e-05 ***
PM_VISIT_LAST_2_YRS 0.324927 0.028819 11.275 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.276194 0.028930 9.547 < 2e-16 ***
AF_25K_GIFT 0.569050 0.043065 13.214 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.259652 0.006808 38.138 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.914343 0.042620 68.379 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.576876 0.025602 22.532 < 2e-16 ***
MG_250K_PLUS 1.071882 0.071627 14.965 < 2e-16 ***
Alumnus -0.570608 0.021474 -26.572 < 2e-16 ***
SEASON_TICKET_YEARS -0.022116 0.004461 -4.958 7.19e-07 ***
AFFINITY_SCORE 0.159975 0.006819 23.460 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.119662 0.022296 -5.367 8.10e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.208362 0.028645 7.274 3.63e-13 ***
MG_PR_MODEL_DESCTop Tier 0.588830 0.028309 20.800 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9655 on 18759 degrees of freedom
Multiple R-squared: 0.7255, Adjusted R-squared: 0.7252
F-statistic: 2754 on 18 and 18759 DF, p-value: < 2.2e-16
[[10]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = 5) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3449 -0.6302 -0.1870 0.5008 6.2783
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.245968 0.093356 2.635 0.00843 **
ns(NUMERIC_AGE, df = 5)1 0.385387 0.085217 4.522 6.15e-06 ***
ns(NUMERIC_AGE, df = 5)2 0.270300 0.099870 2.707 0.00681 **
ns(NUMERIC_AGE, df = 5)3 -0.364235 0.069246 -5.260 1.46e-07 ***
ns(NUMERIC_AGE, df = 5)4 0.538473 0.211455 2.547 0.01089 *
ns(NUMERIC_AGE, df = 5)5 -0.465707 0.114036 -4.084 4.45e-05 ***
PM_VISIT_LAST_2_YRS 0.311610 0.028905 10.781 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.283431 0.029058 9.754 < 2e-16 ***
AF_25K_GIFT 0.612449 0.042930 14.266 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.254815 0.006813 37.400 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.919716 0.042574 68.580 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.582585 0.025656 22.708 < 2e-16 ***
MG_250K_PLUS 1.062557 0.071953 14.767 < 2e-16 ***
Alumnus -0.567552 0.021520 -26.373 < 2e-16 ***
SEASON_TICKET_YEARS -0.021359 0.004514 -4.732 2.24e-06 ***
AFFINITY_SCORE 0.161414 0.006832 23.628 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.112613 0.022355 -5.037 4.76e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.203528 0.028624 7.111 1.20e-12 ***
MG_PR_MODEL_DESCTop Tier 0.577537 0.028417 20.323 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9675 on 18755 degrees of freedom
Multiple R-squared: 0.7231, Adjusted R-squared: 0.7229
F-statistic: 2721 on 18 and 18755 DF, p-value: < 2.2e-16Campaign all predictors splines
lapply(clmaps3
, function(x) {
lapply(x, function(y) summary(y))
}
)[[1]]
[[1]][[1]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3996 -0.6295 -0.1754 0.4992 5.5008
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.011648 0.035936 28.151 < 2e-16 ***
ns(NUMERIC_AGE, df = s) -1.350353 0.074412 -18.147 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.325019 0.028767 11.298 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.272621 0.028949 9.417 < 2e-16 ***
AF_25K_GIFT 0.627879 0.042636 14.727 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.259172 0.006755 38.369 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.951105 0.042358 69.670 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.566090 0.025556 22.151 < 2e-16 ***
MG_250K_PLUS 1.112309 0.071750 15.502 < 2e-16 ***
Alumnus -0.562567 0.020878 -26.945 < 2e-16 ***
SEASON_TICKET_YEARS -0.018714 0.004471 -4.186 2.86e-05 ***
AFFINITY_SCORE 0.151678 0.006755 22.453 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.097397 0.022043 -4.418 1.00e-05 ***
MG_PR_MODEL_DESCMiddle Tier 0.232656 0.028336 8.211 2.34e-16 ***
MG_PR_MODEL_DESCTop Tier 0.616852 0.028159 21.906 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9638 on 18763 degrees of freedom
Multiple R-squared: 0.7255, Adjusted R-squared: 0.7253
F-statistic: 3542 on 14 and 18763 DF, p-value: < 2.2e-16
[[1]][[2]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4424 -0.6269 -0.1756 0.4982 6.3167
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.011224 0.035844 28.212 < 2e-16 ***
ns(NUMERIC_AGE, df = s) -1.303478 0.072326 -18.022 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.313448 0.028825 10.874 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.277762 0.029028 9.569 < 2e-16 ***
AF_25K_GIFT 0.601112 0.043263 13.895 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.257642 0.006743 38.210 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.938189 0.042340 69.395 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.566231 0.025590 22.127 < 2e-16 ***
MG_250K_PLUS 1.155986 0.070177 16.473 < 2e-16 ***
Alumnus -0.559664 0.020901 -26.777 < 2e-16 ***
SEASON_TICKET_YEARS -0.018445 0.004516 -4.085 4.43e-05 ***
AFFINITY_SCORE 0.151976 0.006751 22.510 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.102657 0.022011 -4.664 3.12e-06 ***
MG_PR_MODEL_DESCMiddle Tier 0.226995 0.028326 8.014 1.18e-15 ***
MG_PR_MODEL_DESCTop Tier 0.622891 0.028283 22.023 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9637 on 18763 degrees of freedom
Multiple R-squared: 0.7258, Adjusted R-squared: 0.7256
F-statistic: 3547 on 14 and 18763 DF, p-value: < 2.2e-16
[[1]][[3]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3651 -0.6291 -0.1780 0.4978 6.3608
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.014388 0.035945 28.220 < 2e-16 ***
ns(NUMERIC_AGE, df = s) -1.376220 0.074320 -18.518 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.330153 0.028824 11.454 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.262787 0.028953 9.076 < 2e-16 ***
AF_25K_GIFT 0.624961 0.042805 14.600 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.260261 0.006763 38.483 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.940863 0.042426 69.317 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.571570 0.025680 22.257 < 2e-16 ***
MG_250K_PLUS 1.078021 0.071230 15.134 < 2e-16 ***
Alumnus -0.565383 0.020899 -27.053 < 2e-16 ***
SEASON_TICKET_YEARS -0.019854 0.004429 -4.483 7.41e-06 ***
AFFINITY_SCORE 0.154440 0.006779 22.783 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.091209 0.022082 -4.131 3.63e-05 ***
MG_PR_MODEL_DESCMiddle Tier 0.224616 0.028440 7.898 2.99e-15 ***
MG_PR_MODEL_DESCTop Tier 0.615312 0.028292 21.748 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9665 on 18763 degrees of freedom
Multiple R-squared: 0.7245, Adjusted R-squared: 0.7243
F-statistic: 3525 on 14 and 18763 DF, p-value: < 2.2e-16
[[1]][[4]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3687 -0.6289 -0.1794 0.5053 6.3332
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.025517 0.036033 28.460 < 2e-16 ***
ns(NUMERIC_AGE, df = s) -1.341706 0.074517 -18.005 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.329548 0.029005 11.362 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.287559 0.029000 9.916 < 2e-16 ***
AF_25K_GIFT 0.637148 0.042965 14.829 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.256744 0.006775 37.898 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.951075 0.042396 69.608 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.585934 0.025590 22.897 < 2e-16 ***
MG_250K_PLUS 1.080915 0.071526 15.112 < 2e-16 ***
Alumnus -0.561766 0.020942 -26.825 < 2e-16 ***
SEASON_TICKET_YEARS -0.019398 0.004453 -4.356 1.33e-05 ***
AFFINITY_SCORE 0.149421 0.006789 22.008 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.105363 0.022143 -4.758 1.97e-06 ***
MG_PR_MODEL_DESCMiddle Tier 0.210170 0.028516 7.370 1.77e-13 ***
MG_PR_MODEL_DESCTop Tier 0.597198 0.028339 21.073 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9676 on 18763 degrees of freedom
Multiple R-squared: 0.7228, Adjusted R-squared: 0.7226
F-statistic: 3494 on 14 and 18763 DF, p-value: < 2.2e-16
[[1]][[5]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.1738 -0.6321 -0.1803 0.5071 6.3228
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.011858 0.036129 28.007 < 2e-16 ***
ns(NUMERIC_AGE, df = s) -1.327373 0.074862 -17.731 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.350895 0.029123 12.049 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.275742 0.029204 9.442 < 2e-16 ***
AF_25K_GIFT 0.619291 0.043116 14.363 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.256291 0.006777 37.815 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.941430 0.042618 69.018 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.567490 0.025768 22.023 < 2e-16 ***
MG_250K_PLUS 1.085089 0.072006 15.069 < 2e-16 ***
Alumnus -0.564320 0.020998 -26.875 < 2e-16 ***
SEASON_TICKET_YEARS -0.020180 0.004525 -4.460 8.24e-06 ***
AFFINITY_SCORE 0.152335 0.006804 22.389 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.096627 0.022209 -4.351 1.36e-05 ***
MG_PR_MODEL_DESCMiddle Tier 0.239469 0.028503 8.402 < 2e-16 ***
MG_PR_MODEL_DESCTop Tier 0.615913 0.028366 21.713 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9708 on 18763 degrees of freedom
Multiple R-squared: 0.7215, Adjusted R-squared: 0.7212
F-statistic: 3471 on 14 and 18763 DF, p-value: < 2.2e-16
[[1]][[6]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4019 -0.6267 -0.1800 0.5021 6.3318
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.039185 0.035951 28.905 < 2e-16 ***
ns(NUMERIC_AGE, df = s) -1.369678 0.074403 -18.409 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.296359 0.029183 10.155 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.301114 0.028972 10.393 < 2e-16 ***
AF_25K_GIFT 0.627492 0.043570 14.402 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.257380 0.006770 38.019 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.942504 0.042476 69.275 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.570190 0.025607 22.267 < 2e-16 ***
MG_250K_PLUS 1.125617 0.072414 15.544 < 2e-16 ***
Alumnus -0.568128 0.020836 -27.266 < 2e-16 ***
SEASON_TICKET_YEARS -0.018497 0.004454 -4.153 3.30e-05 ***
AFFINITY_SCORE 0.152032 0.006780 22.422 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.109376 0.022103 -4.948 7.54e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.212158 0.028431 7.462 8.87e-14 ***
MG_PR_MODEL_DESCTop Tier 0.590386 0.028242 20.904 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9656 on 18763 degrees of freedom
Multiple R-squared: 0.7236, Adjusted R-squared: 0.7234
F-statistic: 3509 on 14 and 18763 DF, p-value: < 2.2e-16
[[1]][[7]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3667 -0.6311 -0.1796 0.5034 6.3455
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.011044 0.036031 28.061 < 2e-16 ***
ns(NUMERIC_AGE, df = s) -1.369740 0.074490 -18.388 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.316452 0.028983 10.919 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.298130 0.028976 10.289 < 2e-16 ***
AF_25K_GIFT 0.635174 0.043014 14.767 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.260128 0.006766 38.446 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.924981 0.042463 68.883 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.575521 0.025644 22.443 < 2e-16 ***
MG_250K_PLUS 1.081976 0.072759 14.871 < 2e-16 ***
Alumnus -0.559264 0.020915 -26.740 < 2e-16 ***
SEASON_TICKET_YEARS -0.018541 0.004544 -4.080 4.52e-05 ***
AFFINITY_SCORE 0.151675 0.006784 22.359 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.090844 0.022137 -4.104 4.08e-05 ***
MG_PR_MODEL_DESCMiddle Tier 0.229319 0.028473 8.054 8.49e-16 ***
MG_PR_MODEL_DESCTop Tier 0.599729 0.028290 21.200 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9665 on 18763 degrees of freedom
Multiple R-squared: 0.7231, Adjusted R-squared: 0.7229
F-statistic: 3500 on 14 and 18763 DF, p-value: < 2.2e-16
[[1]][[8]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4171 -0.6331 -0.1760 0.5027 6.3109
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.010263 0.036100 27.985 < 2e-16 ***
ns(NUMERIC_AGE, df = s) -1.320875 0.074626 -17.700 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.329760 0.029233 11.281 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.285037 0.029177 9.769 < 2e-16 ***
AF_25K_GIFT 0.592203 0.043761 13.533 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.256081 0.006812 37.590 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.927337 0.042545 68.805 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.564427 0.025811 21.868 < 2e-16 ***
MG_250K_PLUS 1.119567 0.072456 15.452 < 2e-16 ***
Alumnus -0.559004 0.021015 -26.600 < 2e-16 ***
SEASON_TICKET_YEARS -0.018367 0.004496 -4.085 4.42e-05 ***
AFFINITY_SCORE 0.153385 0.006811 22.521 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.096114 0.022170 -4.335 1.46e-05 ***
MG_PR_MODEL_DESCMiddle Tier 0.219171 0.028544 7.678 1.69e-14 ***
MG_PR_MODEL_DESCTop Tier 0.608056 0.028378 21.427 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9711 on 18763 degrees of freedom
Multiple R-squared: 0.7201, Adjusted R-squared: 0.7199
F-statistic: 3448 on 14 and 18763 DF, p-value: < 2.2e-16
[[1]][[9]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3526 -0.6266 -0.1798 0.4974 6.3268
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.995951 0.036160 27.543 < 2e-16 ***
ns(NUMERIC_AGE, df = s) -1.310639 0.074769 -17.529 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.337107 0.028861 11.680 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.274510 0.029006 9.464 < 2e-16 ***
AF_25K_GIFT 0.582543 0.043143 13.503 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.257403 0.006790 37.909 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.940231 0.042673 68.901 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.586084 0.025660 22.841 < 2e-16 ***
MG_250K_PLUS 1.061446 0.071791 14.785 < 2e-16 ***
Alumnus -0.560707 0.020934 -26.784 < 2e-16 ***
SEASON_TICKET_YEARS -0.021624 0.004472 -4.836 1.34e-06 ***
AFFINITY_SCORE 0.153127 0.006775 22.603 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.097670 0.022199 -4.400 1.09e-05 ***
MG_PR_MODEL_DESCMiddle Tier 0.228182 0.028627 7.971 1.66e-15 ***
MG_PR_MODEL_DESCTop Tier 0.611407 0.028263 21.632 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9683 on 18763 degrees of freedom
Multiple R-squared: 0.7238, Adjusted R-squared: 0.7236
F-statistic: 3512 on 14 and 18763 DF, p-value: < 2.2e-16
[[1]][[10]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3157 -0.6334 -0.1816 0.5059 6.3183
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.000846 0.036132 27.700 < 2e-16 ***
ns(NUMERIC_AGE, df = s) -1.308831 0.074566 -17.553 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.323402 0.028938 11.176 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.281649 0.029129 9.669 < 2e-16 ***
AF_25K_GIFT 0.625439 0.043002 14.545 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.252613 0.006790 37.205 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.944353 0.042620 69.084 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.591298 0.025712 22.997 < 2e-16 ***
MG_250K_PLUS 1.052868 0.072106 14.602 < 2e-16 ***
Alumnus -0.556932 0.020978 -26.548 < 2e-16 ***
SEASON_TICKET_YEARS -0.021093 0.004525 -4.661 3.16e-06 ***
AFFINITY_SCORE 0.154948 0.006792 22.812 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.093206 0.022268 -4.186 2.86e-05 ***
MG_PR_MODEL_DESCMiddle Tier 0.220819 0.028608 7.719 1.23e-14 ***
MG_PR_MODEL_DESCTop Tier 0.598102 0.028373 21.080 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9702 on 18759 degrees of freedom
Multiple R-squared: 0.7215, Adjusted R-squared: 0.7213
F-statistic: 3472 on 14 and 18759 DF, p-value: < 2.2e-16
[[2]]
[[2]][[1]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4259 -0.6249 -0.1741 0.4978 5.4738
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.777184 0.054479 14.266 < 2e-16 ***
ns(NUMERIC_AGE, df = s)1 -0.842164 0.094649 -8.898 < 2e-16 ***
ns(NUMERIC_AGE, df = s)2 -1.071294 0.059850 -17.900 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.315957 0.028786 10.976 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.276454 0.028932 9.555 < 2e-16 ***
AF_25K_GIFT 0.617062 0.042642 14.471 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.256970 0.006760 38.013 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.946377 0.042331 69.604 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.562856 0.025541 22.038 < 2e-16 ***
MG_250K_PLUS 1.128694 0.071747 15.732 < 2e-16 ***
Alumnus -0.549836 0.020979 -26.209 < 2e-16 ***
SEASON_TICKET_YEARS -0.019484 0.004469 -4.359 1.31e-05 ***
AFFINITY_SCORE 0.156518 0.006803 23.009 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.110929 0.022151 -5.008 5.56e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.221203 0.028383 7.794 6.85e-15 ***
MG_PR_MODEL_DESCTop Tier 0.604322 0.028220 21.415 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.963 on 18762 degrees of freedom
Multiple R-squared: 0.726, Adjusted R-squared: 0.7257
F-statistic: 3313 on 15 and 18762 DF, p-value: < 2.2e-16
[[2]][[2]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4673 -0.6227 -0.1746 0.4966 6.4457
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.794233 0.054208 14.652 < 2e-16 ***
ns(NUMERIC_AGE, df = s)1 -0.827844 0.094326 -8.776 < 2e-16 ***
ns(NUMERIC_AGE, df = s)2 -1.002516 0.056302 -17.806 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.305239 0.028845 10.582 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.281952 0.029017 9.717 < 2e-16 ***
AF_25K_GIFT 0.590150 0.043280 13.636 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.255730 0.006747 37.900 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.932488 0.042323 69.289 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.563296 0.025577 22.023 < 2e-16 ***
MG_250K_PLUS 1.170852 0.070181 16.683 < 2e-16 ***
Alumnus -0.547835 0.021003 -26.083 < 2e-16 ***
SEASON_TICKET_YEARS -0.018998 0.004514 -4.209 2.58e-05 ***
AFFINITY_SCORE 0.156398 0.006797 23.009 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.115665 0.022130 -5.227 1.74e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.215982 0.028380 7.610 2.87e-14 ***
MG_PR_MODEL_DESCTop Tier 0.610057 0.028365 21.507 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9629 on 18762 degrees of freedom
Multiple R-squared: 0.7262, Adjusted R-squared: 0.726
F-statistic: 3317 on 15 and 18762 DF, p-value: < 2.2e-16
[[2]][[3]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3871 -0.6252 -0.1771 0.4974 6.4730
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.823498 0.054453 15.123 < 2e-16 ***
ns(NUMERIC_AGE, df = s)1 -0.935844 0.094617 -9.891 < 2e-16 ***
ns(NUMERIC_AGE, df = s)2 -1.049151 0.059789 -17.548 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.323109 0.028847 11.201 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.266102 0.028946 9.193 < 2e-16 ***
AF_25K_GIFT 0.615100 0.042833 14.360 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.258487 0.006770 38.181 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.936436 0.042413 69.234 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.569205 0.025671 22.173 < 2e-16 ***
MG_250K_PLUS 1.091971 0.071254 15.325 < 2e-16 ***
Alumnus -0.554642 0.021014 -26.394 < 2e-16 ***
SEASON_TICKET_YEARS -0.020388 0.004428 -4.604 4.17e-06 ***
AFFINITY_SCORE 0.158344 0.006826 23.196 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.102352 0.022198 -4.611 4.04e-06 ***
MG_PR_MODEL_DESCMiddle Tier 0.215190 0.028496 7.552 4.50e-14 ***
MG_PR_MODEL_DESCTop Tier 0.604325 0.028375 21.298 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.966 on 18762 degrees of freedom
Multiple R-squared: 0.7248, Adjusted R-squared: 0.7246
F-statistic: 3295 on 15 and 18762 DF, p-value: < 2.2e-16
[[2]][[4]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3911 -0.6266 -0.1792 0.5018 6.4486
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.831722 0.054424 15.282 < 2e-16 ***
ns(NUMERIC_AGE, df = s)1 -0.900222 0.094559 -9.520 < 2e-16 ***
ns(NUMERIC_AGE, df = s)2 -1.030774 0.059967 -17.189 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.323220 0.029019 11.138 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.290462 0.028990 10.019 < 2e-16 ***
AF_25K_GIFT 0.627988 0.042984 14.610 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.255054 0.006780 37.618 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.946733 0.042381 69.529 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.583634 0.025580 22.816 < 2e-16 ***
MG_250K_PLUS 1.094495 0.071542 15.299 < 2e-16 ***
Alumnus -0.551041 0.021051 -26.176 < 2e-16 ***
SEASON_TICKET_YEARS -0.019967 0.004452 -4.485 7.34e-06 ***
AFFINITY_SCORE 0.153343 0.006836 22.433 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.116602 0.022256 -5.239 1.63e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.199945 0.028581 6.996 2.73e-12 ***
MG_PR_MODEL_DESCTop Tier 0.585906 0.028422 20.614 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.967 on 18762 degrees of freedom
Multiple R-squared: 0.7231, Adjusted R-squared: 0.7229
F-statistic: 3267 on 15 and 18762 DF, p-value: < 2.2e-16
[[2]][[5]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.1809 -0.6296 -0.1774 0.5016 6.4453
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.807472 0.054715 14.758 < 2e-16 ***
ns(NUMERIC_AGE, df = s)1 -0.871426 0.094932 -9.179 < 2e-16 ***
ns(NUMERIC_AGE, df = s)2 -1.033357 0.060586 -17.056 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.343734 0.029141 11.796 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.278964 0.029192 9.556 < 2e-16 ***
AF_25K_GIFT 0.609923 0.043130 14.142 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.254474 0.006783 37.516 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.937241 0.042599 68.950 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.564534 0.025759 21.916 < 2e-16 ***
MG_250K_PLUS 1.098349 0.072010 15.253 < 2e-16 ***
Alumnus -0.552682 0.021115 -26.175 < 2e-16 ***
SEASON_TICKET_YEARS -0.020814 0.004524 -4.601 4.23e-06 ***
AFFINITY_SCORE 0.156475 0.006851 22.841 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.108758 0.022328 -4.871 1.12e-06 ***
MG_PR_MODEL_DESCMiddle Tier 0.229200 0.028560 8.025 1.07e-15 ***
MG_PR_MODEL_DESCTop Tier 0.604290 0.028444 21.245 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9701 on 18762 degrees of freedom
Multiple R-squared: 0.7218, Adjusted R-squared: 0.7216
F-statistic: 3246 on 15 and 18762 DF, p-value: < 2.2e-16
[[2]][[6]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4267 -0.6229 -0.1770 0.4995 6.4662
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.814980 0.054150 15.050 < 2e-16 ***
ns(NUMERIC_AGE, df = s)1 -0.876078 0.094159 -9.304 < 2e-16 ***
ns(NUMERIC_AGE, df = s)2 -1.078449 0.060033 -17.964 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.288239 0.029196 9.872 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.304469 0.028955 10.515 < 2e-16 ***
AF_25K_GIFT 0.616901 0.043578 14.156 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.255506 0.006773 37.725 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.935560 0.042461 69.136 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.567033 0.025593 22.156 < 2e-16 ***
MG_250K_PLUS 1.139479 0.072400 15.739 < 2e-16 ***
Alumnus -0.556020 0.020934 -26.560 < 2e-16 ***
SEASON_TICKET_YEARS -0.019182 0.004452 -4.308 1.65e-05 ***
AFFINITY_SCORE 0.156750 0.006828 22.955 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.123186 0.022226 -5.542 3.02e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.200280 0.028489 7.030 2.14e-12 ***
MG_PR_MODEL_DESCTop Tier 0.577337 0.028318 20.388 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9648 on 18762 degrees of freedom
Multiple R-squared: 0.7241, Adjusted R-squared: 0.7238
F-statistic: 3282 on 15 and 18762 DF, p-value: < 2.2e-16
[[2]][[7]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3877 -0.6282 -0.1762 0.5003 6.4626
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.817828 0.054304 15.060 < 2e-16 ***
ns(NUMERIC_AGE, df = s)1 -0.927772 0.094183 -9.851 < 2e-16 ***
ns(NUMERIC_AGE, df = s)2 -1.051088 0.060242 -17.448 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.309881 0.028999 10.686 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.300669 0.028964 10.381 < 2e-16 ***
AF_25K_GIFT 0.627178 0.043022 14.578 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.258529 0.006770 38.185 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.920189 0.042451 68.790 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.573120 0.025634 22.358 < 2e-16 ***
MG_250K_PLUS 1.094343 0.072764 15.040 < 2e-16 ***
Alumnus -0.548623 0.021022 -26.097 < 2e-16 ***
SEASON_TICKET_YEARS -0.019103 0.004543 -4.205 2.62e-05 ***
AFFINITY_SCORE 0.155583 0.006829 22.782 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.102004 0.022249 -4.585 4.58e-06 ***
MG_PR_MODEL_DESCMiddle Tier 0.220011 0.028524 7.713 1.29e-14 ***
MG_PR_MODEL_DESCTop Tier 0.588639 0.028369 20.749 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.966 on 18762 degrees of freedom
Multiple R-squared: 0.7234, Adjusted R-squared: 0.7232
F-statistic: 3272 on 15 and 18762 DF, p-value: < 2.2e-16
[[2]][[8]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4421 -0.6277 -0.1758 0.4993 6.4392
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.791753 0.054557 14.512 < 2e-16 ***
ns(NUMERIC_AGE, df = s)1 -0.840441 0.094864 -8.859 < 2e-16 ***
ns(NUMERIC_AGE, df = s)2 -1.038232 0.059933 -17.323 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.321922 0.029248 11.007 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.288644 0.029164 9.897 < 2e-16 ***
AF_25K_GIFT 0.581669 0.043773 13.288 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.254095 0.006818 37.271 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.921771 0.042527 68.704 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.560944 0.025801 21.742 < 2e-16 ***
MG_250K_PLUS 1.134699 0.072459 15.660 < 2e-16 ***
Alumnus -0.546798 0.021124 -25.885 < 2e-16 ***
SEASON_TICKET_YEARS -0.019027 0.004494 -4.234 2.31e-05 ***
AFFINITY_SCORE 0.157941 0.006859 23.026 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.109091 0.022287 -4.895 9.92e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.208235 0.028597 7.282 3.43e-13 ***
MG_PR_MODEL_DESCTop Tier 0.595443 0.028455 20.926 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9704 on 18762 degrees of freedom
Multiple R-squared: 0.7205, Adjusted R-squared: 0.7203
F-statistic: 3225 on 15 and 18762 DF, p-value: < 2.2e-16
[[2]][[9]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3770 -0.6226 -0.1755 0.4943 6.4604
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.773305 0.054446 14.203 < 2e-16 ***
ns(NUMERIC_AGE, df = s)1 -0.825544 0.094585 -8.728 < 2e-16 ***
ns(NUMERIC_AGE, df = s)2 -1.039617 0.060370 -17.221 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.328862 0.028878 11.388 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.277902 0.028990 9.586 < 2e-16 ***
AF_25K_GIFT 0.572696 0.043148 13.273 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.255514 0.006794 37.611 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.933427 0.042659 68.765 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.582673 0.025647 22.719 < 2e-16 ***
MG_250K_PLUS 1.075064 0.071779 14.977 < 2e-16 ***
Alumnus -0.548794 0.021032 -26.094 < 2e-16 ***
SEASON_TICKET_YEARS -0.022380 0.004470 -5.006 5.60e-07 ***
AFFINITY_SCORE 0.157850 0.006824 23.131 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.111074 0.022318 -4.977 6.52e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.216703 0.028682 7.555 4.36e-14 ***
MG_PR_MODEL_DESCTop Tier 0.598574 0.028339 21.122 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9675 on 18762 degrees of freedom
Multiple R-squared: 0.7242, Adjusted R-squared: 0.724
F-statistic: 3285 on 15 and 18762 DF, p-value: < 2.2e-16
[[2]][[10]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3377 -0.6311 -0.1801 0.5051 6.4433
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.790300 0.054442 14.516 < 2e-16 ***
ns(NUMERIC_AGE, df = s)1 -0.843606 0.094468 -8.930 < 2e-16 ***
ns(NUMERIC_AGE, df = s)2 -1.025463 0.060041 -17.079 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.315214 0.028961 10.884 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.285010 0.029116 9.789 < 2e-16 ***
AF_25K_GIFT 0.616050 0.043010 14.323 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.250747 0.006795 36.903 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.939793 0.042600 69.010 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.588759 0.025699 22.910 < 2e-16 ***
MG_250K_PLUS 1.066056 0.072102 14.785 < 2e-16 ***
Alumnus -0.545074 0.021089 -25.847 < 2e-16 ***
SEASON_TICKET_YEARS -0.021622 0.004523 -4.780 1.76e-06 ***
AFFINITY_SCORE 0.159183 0.006837 23.282 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.105552 0.022381 -4.716 2.42e-06 ***
MG_PR_MODEL_DESCMiddle Tier 0.210380 0.028660 7.341 2.21e-13 ***
MG_PR_MODEL_DESCTop Tier 0.586396 0.028444 20.616 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9696 on 18758 degrees of freedom
Multiple R-squared: 0.7219, Adjusted R-squared: 0.7217
F-statistic: 3247 on 15 and 18758 DF, p-value: < 2.2e-16
[[3]]
[[3]][[1]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4360 -0.6219 -0.1800 0.4941 5.4672
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.358075 0.072991 4.906 9.39e-07 ***
ns(NUMERIC_AGE, df = s)1 -0.452759 0.044995 -10.062 < 2e-16 ***
ns(NUMERIC_AGE, df = s)2 0.282335 0.154973 1.822 0.0685 .
ns(NUMERIC_AGE, df = s)3 -0.522346 0.087683 -5.957 2.61e-09 ***
PM_VISIT_LAST_2_YRS 0.312085 0.028734 10.861 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.274577 0.028877 9.509 < 2e-16 ***
AF_25K_GIFT 0.614415 0.042560 14.436 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.261445 0.006767 38.636 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.929279 0.042295 69.258 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.557954 0.025497 21.883 < 2e-16 ***
MG_250K_PLUS 1.126097 0.071608 15.726 < 2e-16 ***
Alumnus -0.567458 0.021038 -26.973 < 2e-16 ***
SEASON_TICKET_YEARS -0.019120 0.004461 -4.286 1.83e-05 ***
AFFINITY_SCORE 0.158268 0.006792 23.301 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.118449 0.022125 -5.353 8.73e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.213531 0.028342 7.534 5.14e-14 ***
MG_PR_MODEL_DESCTop Tier 0.594237 0.028189 21.080 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9611 on 18761 degrees of freedom
Multiple R-squared: 0.727, Adjusted R-squared: 0.7268
F-statistic: 3123 on 16 and 18761 DF, p-value: < 2.2e-16
[[3]][[2]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4752 -0.6194 -0.1824 0.4939 6.2696
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.360563 0.073273 4.921 8.69e-07 ***
ns(NUMERIC_AGE, df = s)1 -0.429283 0.044473 -9.653 < 2e-16 ***
ns(NUMERIC_AGE, df = s)2 0.317415 0.154324 2.057 0.0397 *
ns(NUMERIC_AGE, df = s)3 -0.501935 0.080845 -6.209 5.46e-10 ***
PM_VISIT_LAST_2_YRS 0.302292 0.028789 10.500 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.279324 0.028960 9.645 < 2e-16 ***
AF_25K_GIFT 0.587303 0.043194 13.597 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.259958 0.006751 38.506 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.913936 0.042290 68.903 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.557904 0.025533 21.850 < 2e-16 ***
MG_250K_PLUS 1.167391 0.070040 16.668 < 2e-16 ***
Alumnus -0.567248 0.021076 -26.914 < 2e-16 ***
SEASON_TICKET_YEARS -0.018659 0.004505 -4.142 3.46e-05 ***
AFFINITY_SCORE 0.158445 0.006788 23.344 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.123704 0.022104 -5.596 2.22e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.207412 0.028340 7.319 2.60e-13 ***
MG_PR_MODEL_DESCTop Tier 0.599301 0.028334 21.151 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.961 on 18761 degrees of freedom
Multiple R-squared: 0.7273, Adjusted R-squared: 0.7271
F-statistic: 3127 on 16 and 18761 DF, p-value: < 2.2e-16
[[3]][[3]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3990 -0.6241 -0.1847 0.4941 6.3025
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.381157 0.074184 5.138 2.81e-07 ***
ns(NUMERIC_AGE, df = s)1 -0.478745 0.045335 -10.560 < 2e-16 ***
ns(NUMERIC_AGE, df = s)2 0.251596 0.157475 1.598 0.11
ns(NUMERIC_AGE, df = s)3 -0.513003 0.086761 -5.913 3.42e-09 ***
PM_VISIT_LAST_2_YRS 0.320032 0.028791 11.116 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.263114 0.028890 9.108 < 2e-16 ***
AF_25K_GIFT 0.613361 0.042747 14.348 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.263096 0.006777 38.823 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.916887 0.042387 68.816 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.563482 0.025628 21.987 < 2e-16 ***
MG_250K_PLUS 1.090068 0.071111 15.329 < 2e-16 ***
Alumnus -0.573975 0.021087 -27.220 < 2e-16 ***
SEASON_TICKET_YEARS -0.020058 0.004419 -4.539 5.69e-06 ***
AFFINITY_SCORE 0.160298 0.006816 23.517 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.110686 0.022174 -4.992 6.04e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.206903 0.028454 7.271 3.70e-13 ***
MG_PR_MODEL_DESCTop Tier 0.594414 0.028340 20.974 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.964 on 18761 degrees of freedom
Multiple R-squared: 0.7259, Adjusted R-squared: 0.7257
F-statistic: 3106 on 16 and 18761 DF, p-value: < 2.2e-16
[[3]][[4]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4057 -0.6229 -0.1866 0.4963 6.2590
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.378194 0.072866 5.190 2.12e-07 ***
ns(NUMERIC_AGE, df = s)1 -0.492308 0.045010 -10.938 < 2e-16 ***
ns(NUMERIC_AGE, df = s)2 0.326394 0.154835 2.108 0.035 *
ns(NUMERIC_AGE, df = s)3 -0.443501 0.087746 -5.054 4.36e-07 ***
PM_VISIT_LAST_2_YRS 0.319655 0.028955 11.040 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.288018 0.028925 9.957 < 2e-16 ***
AF_25K_GIFT 0.624199 0.042887 14.554 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.259949 0.006785 38.313 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.925810 0.042344 69.097 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.578616 0.025527 22.667 < 2e-16 ***
MG_250K_PLUS 1.093725 0.071378 15.323 < 2e-16 ***
Alumnus -0.570607 0.021107 -27.034 < 2e-16 ***
SEASON_TICKET_YEARS -0.019557 0.004442 -4.403 1.08e-05 ***
AFFINITY_SCORE 0.155470 0.006824 22.784 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.124777 0.022222 -5.615 1.99e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.191738 0.028529 6.721 1.86e-11 ***
MG_PR_MODEL_DESCTop Tier 0.575477 0.028379 20.278 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9648 on 18761 degrees of freedom
Multiple R-squared: 0.7244, Adjusted R-squared: 0.7242
F-statistic: 3082 on 16 and 18761 DF, p-value: < 2.2e-16
[[3]][[5]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.1005 -0.6261 -0.1846 0.4979 6.2612
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.376616 0.073279 5.139 2.78e-07 ***
ns(NUMERIC_AGE, df = s)1 -0.472476 0.045194 -10.454 < 2e-16 ***
ns(NUMERIC_AGE, df = s)2 0.294999 0.155954 1.892 0.0586 .
ns(NUMERIC_AGE, df = s)3 -0.467416 0.088882 -5.259 1.47e-07 ***
PM_VISIT_LAST_2_YRS 0.339512 0.029085 11.673 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.277188 0.029134 9.514 < 2e-16 ***
AF_25K_GIFT 0.606429 0.043044 14.089 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.259101 0.006790 38.162 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.917672 0.042570 68.537 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.559171 0.025713 21.746 < 2e-16 ***
MG_250K_PLUS 1.097861 0.071863 15.277 < 2e-16 ***
Alumnus -0.571250 0.021177 -26.976 < 2e-16 ***
SEASON_TICKET_YEARS -0.020725 0.004514 -4.591 4.44e-06 ***
AFFINITY_SCORE 0.158559 0.006841 23.179 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.116854 0.022302 -5.240 1.63e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.220258 0.028520 7.723 1.19e-14 ***
MG_PR_MODEL_DESCTop Tier 0.593793 0.028411 20.900 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9682 on 18761 degrees of freedom
Multiple R-squared: 0.723, Adjusted R-squared: 0.7227
F-statistic: 3060 on 16 and 18761 DF, p-value: < 2.2e-16
[[3]][[6]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4436 -0.6202 -0.1833 0.4942 6.2679
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.346070 0.072294 4.787 1.71e-06 ***
ns(NUMERIC_AGE, df = s)1 -0.485810 0.044860 -10.830 < 2e-16 ***
ns(NUMERIC_AGE, df = s)2 0.393332 0.154097 2.552 0.0107 *
ns(NUMERIC_AGE, df = s)3 -0.457639 0.087711 -5.218 1.83e-07 ***
PM_VISIT_LAST_2_YRS 0.284077 0.029127 9.753 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.301108 0.028885 10.424 < 2e-16 ***
AF_25K_GIFT 0.613375 0.043470 14.110 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.260695 0.006777 38.469 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.912663 0.042419 68.664 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.561869 0.025535 22.004 < 2e-16 ***
MG_250K_PLUS 1.141220 0.072219 15.802 < 2e-16 ***
Alumnus -0.576644 0.020988 -27.475 < 2e-16 ***
SEASON_TICKET_YEARS -0.018921 0.004441 -4.260 2.05e-05 ***
AFFINITY_SCORE 0.158847 0.006815 23.309 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.133122 0.022194 -5.998 2.03e-09 ***
MG_PR_MODEL_DESCMiddle Tier 0.190989 0.028434 6.717 1.91e-11 ***
MG_PR_MODEL_DESCTop Tier 0.565182 0.028275 19.989 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9624 on 18761 degrees of freedom
Multiple R-squared: 0.7254, Adjusted R-squared: 0.7252
F-statistic: 3098 on 16 and 18761 DF, p-value: < 2.2e-16
[[3]][[7]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4000 -0.6259 -0.1864 0.4954 6.2695
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.348360 0.073295 4.753 2.02e-06 ***
ns(NUMERIC_AGE, df = s)1 -0.489848 0.045300 -10.813 < 2e-16 ***
ns(NUMERIC_AGE, df = s)2 0.342897 0.155895 2.200 0.0279 *
ns(NUMERIC_AGE, df = s)3 -0.458526 0.087544 -5.238 1.64e-07 ***
PM_VISIT_LAST_2_YRS 0.305501 0.028934 10.559 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.299215 0.028896 10.355 < 2e-16 ***
AF_25K_GIFT 0.623214 0.042922 14.520 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.263296 0.006773 38.874 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.899372 0.042406 68.371 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.567615 0.025580 22.190 < 2e-16 ***
MG_250K_PLUS 1.093032 0.072591 15.057 < 2e-16 ***
Alumnus -0.570306 0.021095 -27.035 < 2e-16 ***
SEASON_TICKET_YEARS -0.018910 0.004532 -4.172 3.03e-05 ***
AFFINITY_SCORE 0.157740 0.006817 23.140 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.110059 0.022212 -4.955 7.30e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.211672 0.028469 7.435 1.09e-13 ***
MG_PR_MODEL_DESCTop Tier 0.577682 0.028325 20.395 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9637 on 18761 degrees of freedom
Multiple R-squared: 0.7248, Adjusted R-squared: 0.7245
F-statistic: 3088 on 16 and 18761 DF, p-value: < 2.2e-16
[[3]][[8]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4509 -0.6266 -0.1805 0.4972 6.2585
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.328595 0.073978 4.442 8.97e-06 ***
ns(NUMERIC_AGE, df = s)1 -0.442751 0.045512 -9.728 < 2e-16 ***
ns(NUMERIC_AGE, df = s)2 0.396283 0.157093 2.523 0.0117 *
ns(NUMERIC_AGE, df = s)3 -0.460994 0.086824 -5.310 1.11e-07 ***
PM_VISIT_LAST_2_YRS 0.318523 0.029185 10.914 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.286935 0.029099 9.861 < 2e-16 ***
AF_25K_GIFT 0.576223 0.043679 13.192 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.259068 0.006824 37.967 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.901589 0.042487 68.293 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.555584 0.025749 21.577 < 2e-16 ***
MG_250K_PLUS 1.128282 0.072299 15.606 < 2e-16 ***
Alumnus -0.567554 0.021195 -26.778 < 2e-16 ***
SEASON_TICKET_YEARS -0.018724 0.004484 -4.175 2.99e-05 ***
AFFINITY_SCORE 0.159774 0.006847 23.336 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.117972 0.022257 -5.300 1.17e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.199455 0.028548 6.987 2.91e-12 ***
MG_PR_MODEL_DESCTop Tier 0.584461 0.028416 20.568 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9682 on 18761 degrees of freedom
Multiple R-squared: 0.7218, Adjusted R-squared: 0.7216
F-statistic: 3042 on 16 and 18761 DF, p-value: < 2.2e-16
[[3]][[9]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3881 -0.6208 -0.1851 0.4939 6.2782
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.321066 0.073758 4.353 1.35e-05 ***
ns(NUMERIC_AGE, df = s)1 -0.433469 0.045387 -9.551 < 2e-16 ***
ns(NUMERIC_AGE, df = s)2 0.383908 0.156806 2.448 0.0144 *
ns(NUMERIC_AGE, df = s)3 -0.467241 0.087481 -5.341 9.35e-08 ***
PM_VISIT_LAST_2_YRS 0.325143 0.028819 11.282 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.276145 0.028928 9.546 < 2e-16 ***
AF_25K_GIFT 0.568122 0.043058 13.194 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.260068 0.006798 38.259 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.914326 0.042619 68.381 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.577044 0.025600 22.541 < 2e-16 ***
MG_250K_PLUS 1.072515 0.071624 14.974 < 2e-16 ***
Alumnus -0.568569 0.021099 -26.948 < 2e-16 ***
SEASON_TICKET_YEARS -0.022066 0.004461 -4.947 7.62e-07 ***
AFFINITY_SCORE 0.159860 0.006813 23.464 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.120211 0.022292 -5.393 7.03e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.207551 0.028637 7.248 4.41e-13 ***
MG_PR_MODEL_DESCTop Tier 0.588070 0.028302 20.779 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9655 on 18761 degrees of freedom
Multiple R-squared: 0.7254, Adjusted R-squared: 0.7252
F-statistic: 3098 on 16 and 18761 DF, p-value: < 2.2e-16
[[3]][[10]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3493 -0.6303 -0.1879 0.4979 6.2689
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.351584 0.073557 4.780 1.77e-06 ***
ns(NUMERIC_AGE, df = s)1 -0.442558 0.045391 -9.750 < 2e-16 ***
ns(NUMERIC_AGE, df = s)2 0.331759 0.155961 2.127 0.0334 *
ns(NUMERIC_AGE, df = s)3 -0.476504 0.086788 -5.490 4.06e-08 ***
PM_VISIT_LAST_2_YRS 0.311911 0.028904 10.791 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.283395 0.029057 9.753 < 2e-16 ***
AF_25K_GIFT 0.611193 0.042926 14.238 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.255408 0.006801 37.553 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.919516 0.042574 68.575 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.582768 0.025655 22.715 < 2e-16 ***
MG_250K_PLUS 1.063065 0.071955 14.774 < 2e-16 ***
Alumnus -0.564946 0.021164 -26.693 < 2e-16 ***
SEASON_TICKET_YEARS -0.021312 0.004514 -4.721 2.36e-06 ***
AFFINITY_SCORE 0.161271 0.006827 23.622 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.113356 0.022352 -5.071 3.99e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.202266 0.028615 7.068 1.62e-12 ***
MG_PR_MODEL_DESCTop Tier 0.576232 0.028409 20.284 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9676 on 18757 degrees of freedom
Multiple R-squared: 0.7231, Adjusted R-squared: 0.7228
F-statistic: 3061 on 16 and 18757 DF, p-value: < 2.2e-16
[[4]]
[[4]][[1]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4296 -0.6231 -0.1840 0.4961 5.4705
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.288923 0.082811 3.489 0.000486 ***
ns(NUMERIC_AGE, df = s)1 0.288136 0.078848 3.654 0.000259 ***
ns(NUMERIC_AGE, df = s)2 -0.379694 0.062161 -6.108 1.03e-09 ***
ns(NUMERIC_AGE, df = s)3 0.417619 0.181241 2.304 0.021221 *
ns(NUMERIC_AGE, df = s)4 -0.550520 0.101193 -5.440 5.39e-08 ***
PM_VISIT_LAST_2_YRS 0.311668 0.028734 10.847 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.274296 0.028877 9.499 < 2e-16 ***
AF_25K_GIFT 0.615566 0.042565 14.462 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.260932 0.006777 38.503 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.929497 0.042296 69.261 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.558140 0.025498 21.889 < 2e-16 ***
MG_250K_PLUS 1.125450 0.071607 15.717 < 2e-16 ***
Alumnus -0.571979 0.021385 -26.747 < 2e-16 ***
SEASON_TICKET_YEARS -0.019193 0.004461 -4.302 1.70e-05 ***
AFFINITY_SCORE 0.158169 0.006795 23.279 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.118092 0.022125 -5.338 9.53e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.214319 0.028343 7.562 4.17e-14 ***
MG_PR_MODEL_DESCTop Tier 0.594900 0.028190 21.103 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9611 on 18760 degrees of freedom
Multiple R-squared: 0.7271, Adjusted R-squared: 0.7268
F-statistic: 2940 on 17 and 18760 DF, p-value: < 2.2e-16
[[4]][[2]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4732 -0.6200 -0.1825 0.4953 6.2757
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.323824 0.081663 3.965 7.36e-05 ***
ns(NUMERIC_AGE, df = s)1 0.252712 0.077499 3.261 0.00111 **
ns(NUMERIC_AGE, df = s)2 -0.396870 0.061448 -6.459 1.08e-10 ***
ns(NUMERIC_AGE, df = s)3 0.378713 0.178456 2.122 0.03384 *
ns(NUMERIC_AGE, df = s)4 -0.497885 0.094029 -5.295 1.20e-07 ***
PM_VISIT_LAST_2_YRS 0.302180 0.028790 10.496 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.279260 0.028961 9.643 < 2e-16 ***
AF_25K_GIFT 0.587723 0.043197 13.606 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.259750 0.006764 38.404 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.913928 0.042296 68.894 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.557857 0.025536 21.846 < 2e-16 ***
MG_250K_PLUS 1.167091 0.070044 16.662 < 2e-16 ***
Alumnus -0.568612 0.021419 -26.547 < 2e-16 ***
SEASON_TICKET_YEARS -0.018688 0.004505 -4.148 3.36e-05 ***
AFFINITY_SCORE 0.158480 0.006790 23.340 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.123524 0.022104 -5.588 2.33e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.207792 0.028342 7.332 2.37e-13 ***
MG_PR_MODEL_DESCTop Tier 0.599653 0.028335 21.163 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.961 on 18760 degrees of freedom
Multiple R-squared: 0.7273, Adjusted R-squared: 0.7271
F-statistic: 2943 on 17 and 18760 DF, p-value: < 2.2e-16
[[4]][[3]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3985 -0.6242 -0.1847 0.4942 6.3032
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.354833 0.083365 4.256 2.09e-05 ***
ns(NUMERIC_AGE, df = s)1 0.220002 0.079432 2.770 0.00562 **
ns(NUMERIC_AGE, df = s)2 -0.458367 0.062425 -7.343 2.18e-13 ***
ns(NUMERIC_AGE, df = s)3 0.289234 0.182753 1.583 0.11352
ns(NUMERIC_AGE, df = s)4 -0.495662 0.100754 -4.920 8.75e-07 ***
PM_VISIT_LAST_2_YRS 0.319973 0.028793 11.113 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.263120 0.028892 9.107 < 2e-16 ***
AF_25K_GIFT 0.613658 0.042755 14.353 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.263002 0.006788 38.745 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.916600 0.042391 68.803 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.563353 0.025631 21.980 < 2e-16 ***
MG_250K_PLUS 1.089883 0.071115 15.326 < 2e-16 ***
Alumnus -0.574169 0.021431 -26.791 < 2e-16 ***
SEASON_TICKET_YEARS -0.020072 0.004419 -4.542 5.61e-06 ***
AFFINITY_SCORE 0.160376 0.006819 23.518 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.110567 0.022175 -4.986 6.21e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.207151 0.028457 7.279 3.48e-13 ***
MG_PR_MODEL_DESCTop Tier 0.594695 0.028342 20.983 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.964 on 18760 degrees of freedom
Multiple R-squared: 0.726, Adjusted R-squared: 0.7257
F-statistic: 2923 on 17 and 18760 DF, p-value: < 2.2e-16
[[4]][[4]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4017 -0.6248 -0.1849 0.4958 6.2695
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.325034 0.082395 3.945 8.01e-05 ***
ns(NUMERIC_AGE, df = s)1 0.261160 0.078116 3.343 0.00083 ***
ns(NUMERIC_AGE, df = s)2 -0.430002 0.062840 -6.843 8.00e-12 ***
ns(NUMERIC_AGE, df = s)3 0.425571 0.180091 2.363 0.01813 *
ns(NUMERIC_AGE, df = s)4 -0.447974 0.102477 -4.371 1.24e-05 ***
PM_VISIT_LAST_2_YRS 0.319487 0.028956 11.034 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.287788 0.028927 9.949 < 2e-16 ***
AF_25K_GIFT 0.625000 0.042894 14.571 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.259612 0.006795 38.204 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.925819 0.042346 69.093 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.578604 0.025528 22.665 < 2e-16 ***
MG_250K_PLUS 1.092912 0.071386 15.310 < 2e-16 ***
Alumnus -0.573106 0.021456 -26.710 < 2e-16 ***
SEASON_TICKET_YEARS -0.019603 0.004442 -4.413 1.03e-05 ***
AFFINITY_SCORE 0.155503 0.006825 22.783 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.124479 0.022222 -5.602 2.15e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.192303 0.028530 6.740 1.62e-11 ***
MG_PR_MODEL_DESCTop Tier 0.576021 0.028380 20.297 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9648 on 18760 degrees of freedom
Multiple R-squared: 0.7244, Adjusted R-squared: 0.7242
F-statistic: 2901 on 17 and 18760 DF, p-value: < 2.2e-16
[[4]][[5]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.1051 -0.6283 -0.1865 0.5004 6.2701
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.329943 0.082904 3.980 6.92e-05 ***
ns(NUMERIC_AGE, df = s)1 0.249292 0.078665 3.169 0.00153 **
ns(NUMERIC_AGE, df = s)2 -0.419939 0.063256 -6.639 3.25e-11 ***
ns(NUMERIC_AGE, df = s)3 0.376821 0.181407 2.077 0.03780 *
ns(NUMERIC_AGE, df = s)4 -0.469290 0.103865 -4.518 6.27e-06 ***
PM_VISIT_LAST_2_YRS 0.339327 0.029087 11.666 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.277025 0.029135 9.508 < 2e-16 ***
AF_25K_GIFT 0.607102 0.043050 14.102 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.258789 0.006802 38.049 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.917760 0.042572 68.537 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.559183 0.025716 21.745 < 2e-16 ***
MG_250K_PLUS 1.097380 0.071869 15.269 < 2e-16 ***
Alumnus -0.573278 0.021539 -26.616 < 2e-16 ***
SEASON_TICKET_YEARS -0.020749 0.004514 -4.596 4.33e-06 ***
AFFINITY_SCORE 0.158588 0.006842 23.177 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.116625 0.022302 -5.229 1.72e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.220733 0.028521 7.739 1.05e-14 ***
MG_PR_MODEL_DESCTop Tier 0.594226 0.028412 20.915 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9682 on 18760 degrees of freedom
Multiple R-squared: 0.723, Adjusted R-squared: 0.7227
F-statistic: 2880 on 17 and 18760 DF, p-value: < 2.2e-16
[[4]][[6]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4377 -0.6197 -0.1862 0.4938 6.2851
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.284823 0.081713 3.486 0.000492 ***
ns(NUMERIC_AGE, df = s)1 0.308956 0.077609 3.981 6.89e-05 ***
ns(NUMERIC_AGE, df = s)2 -0.409983 0.062611 -6.548 5.98e-11 ***
ns(NUMERIC_AGE, df = s)3 0.507758 0.179007 2.837 0.004566 **
ns(NUMERIC_AGE, df = s)4 -0.478039 0.102672 -4.656 3.25e-06 ***
PM_VISIT_LAST_2_YRS 0.283734 0.029128 9.741 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.300959 0.028885 10.419 < 2e-16 ***
AF_25K_GIFT 0.614436 0.043477 14.133 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.260220 0.006787 38.342 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.913102 0.042423 68.667 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.562084 0.025537 22.010 < 2e-16 ***
MG_250K_PLUS 1.140005 0.072229 15.783 < 2e-16 ***
Alumnus -0.580115 0.021340 -27.184 < 2e-16 ***
SEASON_TICKET_YEARS -0.018982 0.004442 -4.274 1.93e-05 ***
AFFINITY_SCORE 0.158786 0.006818 23.291 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.132685 0.022194 -5.978 2.29e-09 ***
MG_PR_MODEL_DESCMiddle Tier 0.191778 0.028436 6.744 1.58e-11 ***
MG_PR_MODEL_DESCTop Tier 0.565939 0.028276 20.015 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9624 on 18760 degrees of freedom
Multiple R-squared: 0.7255, Adjusted R-squared: 0.7252
F-statistic: 2916 on 17 and 18760 DF, p-value: < 2.2e-16
[[4]][[7]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4001 -0.6257 -0.1871 0.4949 6.2676
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.322857 0.081648 3.954 7.71e-05 ***
ns(NUMERIC_AGE, df = s)1 0.242376 0.077371 3.133 0.00174 **
ns(NUMERIC_AGE, df = s)2 -0.463822 0.062721 -7.395 1.47e-13 ***
ns(NUMERIC_AGE, df = s)3 0.379954 0.178521 2.128 0.03332 *
ns(NUMERIC_AGE, df = s)4 -0.436114 0.103878 -4.198 2.70e-05 ***
PM_VISIT_LAST_2_YRS 0.305488 0.028935 10.558 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.299286 0.028897 10.357 < 2e-16 ***
AF_25K_GIFT 0.623444 0.042928 14.523 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.263227 0.006783 38.806 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.899179 0.042410 68.361 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.567471 0.025584 22.181 < 2e-16 ***
MG_250K_PLUS 1.092909 0.072594 15.055 < 2e-16 ***
Alumnus -0.570159 0.021442 -26.590 < 2e-16 ***
SEASON_TICKET_YEARS -0.018921 0.004532 -4.175 3.00e-05 ***
AFFINITY_SCORE 0.157842 0.006820 23.144 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.109932 0.022213 -4.949 7.52e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.211905 0.028473 7.442 1.03e-13 ***
MG_PR_MODEL_DESCTop Tier 0.577901 0.028327 20.401 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9637 on 18760 degrees of freedom
Multiple R-squared: 0.7248, Adjusted R-squared: 0.7245
F-statistic: 2906 on 17 and 18760 DF, p-value: < 2.2e-16
[[4]][[8]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4496 -0.6251 -0.1805 0.4962 6.2614
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.294159 0.082738 3.555 0.000378 ***
ns(NUMERIC_AGE, df = s)1 0.288964 0.078520 3.680 0.000234 ***
ns(NUMERIC_AGE, df = s)2 -0.409295 0.063072 -6.489 8.84e-11 ***
ns(NUMERIC_AGE, df = s)3 0.451616 0.181119 2.493 0.012658 *
ns(NUMERIC_AGE, df = s)4 -0.449143 0.102079 -4.400 1.09e-05 ***
PM_VISIT_LAST_2_YRS 0.318483 0.029185 10.913 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.286940 0.029100 9.860 < 2e-16 ***
AF_25K_GIFT 0.576557 0.043684 13.198 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.258906 0.006835 37.880 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.901343 0.042490 68.283 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.555492 0.025751 21.572 < 2e-16 ***
MG_250K_PLUS 1.127942 0.072302 15.600 < 2e-16 ***
Alumnus -0.568424 0.021556 -26.370 < 2e-16 ***
SEASON_TICKET_YEARS -0.018739 0.004485 -4.178 2.95e-05 ***
AFFINITY_SCORE 0.159856 0.006848 23.342 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.117802 0.022258 -5.293 1.22e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.199750 0.028550 6.997 2.71e-12 ***
MG_PR_MODEL_DESCTop Tier 0.584714 0.028417 20.576 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9682 on 18760 degrees of freedom
Multiple R-squared: 0.7218, Adjusted R-squared: 0.7216
F-statistic: 2863 on 17 and 18760 DF, p-value: < 2.2e-16
[[4]][[9]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3855 -0.6218 -0.1867 0.4930 6.2866
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.280355 0.082526 3.397 0.000682 ***
ns(NUMERIC_AGE, df = s)1 0.299286 0.078611 3.807 0.000141 ***
ns(NUMERIC_AGE, df = s)2 -0.393943 0.062216 -6.332 2.48e-10 ***
ns(NUMERIC_AGE, df = s)3 0.454727 0.180833 2.515 0.011924 *
ns(NUMERIC_AGE, df = s)4 -0.465830 0.102074 -4.564 5.06e-06 ***
PM_VISIT_LAST_2_YRS 0.324971 0.028819 11.276 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.276117 0.028929 9.545 < 2e-16 ***
AF_25K_GIFT 0.568679 0.043064 13.206 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.259833 0.006809 38.162 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.914313 0.042623 68.374 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.576984 0.025602 22.537 < 2e-16 ***
MG_250K_PLUS 1.072120 0.071627 14.968 < 2e-16 ***
Alumnus -0.570335 0.021460 -26.577 < 2e-16 ***
SEASON_TICKET_YEARS -0.022090 0.004461 -4.952 7.41e-07 ***
AFFINITY_SCORE 0.159883 0.006815 23.460 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.119966 0.022293 -5.381 7.48e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.207905 0.028638 7.260 4.03e-13 ***
MG_PR_MODEL_DESCTop Tier 0.588369 0.028301 20.790 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9655 on 18760 degrees of freedom
Multiple R-squared: 0.7254, Adjusted R-squared: 0.7252
F-statistic: 2916 on 17 and 18760 DF, p-value: < 2.2e-16
[[4]][[10]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3449 -0.6305 -0.1886 0.5002 6.2836
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.299770 0.081998 3.656 0.000257 ***
ns(NUMERIC_AGE, df = s)1 0.270641 0.077680 3.484 0.000495 ***
ns(NUMERIC_AGE, df = s)2 -0.386445 0.062683 -6.165 7.19e-10 ***
ns(NUMERIC_AGE, df = s)3 0.431885 0.179201 2.410 0.015960 *
ns(NUMERIC_AGE, df = s)4 -0.495068 0.102271 -4.841 1.30e-06 ***
PM_VISIT_LAST_2_YRS 0.311617 0.028905 10.781 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.283248 0.029057 9.748 < 2e-16 ***
AF_25K_GIFT 0.611953 0.042930 14.255 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.255020 0.006813 37.430 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.919673 0.042576 68.575 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.582794 0.025656 22.715 < 2e-16 ***
MG_250K_PLUS 1.062659 0.071954 14.769 < 2e-16 ***
Alumnus -0.568056 0.021513 -26.406 < 2e-16 ***
SEASON_TICKET_YEARS -0.021336 0.004514 -4.727 2.30e-06 ***
AFFINITY_SCORE 0.161245 0.006829 23.612 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.113029 0.022352 -5.057 4.31e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.202817 0.028616 7.087 1.41e-12 ***
MG_PR_MODEL_DESCTop Tier 0.576788 0.028410 20.302 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9676 on 18756 degrees of freedom
Multiple R-squared: 0.7231, Adjusted R-squared: 0.7229
F-statistic: 2881 on 17 and 18756 DF, p-value: < 2.2e-16
[[5]]
[[5]][[1]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4291 -0.6237 -0.1842 0.4958 5.4709
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.255115 0.094156 2.709 0.00675 **
ns(NUMERIC_AGE, df = s)1 0.386056 0.086031 4.487 7.25e-06 ***
ns(NUMERIC_AGE, df = s)2 0.250456 0.100715 2.487 0.01290 *
ns(NUMERIC_AGE, df = s)3 -0.347147 0.069289 -5.010 5.49e-07 ***
ns(NUMERIC_AGE, df = s)4 0.464897 0.213109 2.181 0.02916 *
ns(NUMERIC_AGE, df = s)5 -0.546338 0.114141 -4.787 1.71e-06 ***
PM_VISIT_LAST_2_YRS 0.311628 0.028735 10.845 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.274347 0.028878 9.500 < 2e-16 ***
AF_25K_GIFT 0.615943 0.042566 14.470 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.260803 0.006777 38.485 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.929400 0.042294 69.262 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.558071 0.025498 21.887 < 2e-16 ***
MG_250K_PLUS 1.125296 0.071607 15.715 < 2e-16 ***
Alumnus -0.572103 0.021390 -26.746 < 2e-16 ***
SEASON_TICKET_YEARS -0.019221 0.004461 -4.309 1.65e-05 ***
AFFINITY_SCORE 0.158247 0.006798 23.279 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.117862 0.022128 -5.326 1.01e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.214713 0.028350 7.574 3.80e-14 ***
MG_PR_MODEL_DESCTop Tier 0.595287 0.028195 21.113 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9611 on 18759 degrees of freedom
Multiple R-squared: 0.7271, Adjusted R-squared: 0.7268
F-statistic: 2776 on 18 and 18759 DF, p-value: < 2.2e-16
[[5]][[2]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4727 -0.6214 -0.1825 0.4944 6.2743
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.283317 0.093030 3.045 0.00233 **
ns(NUMERIC_AGE, df = s)1 0.359258 0.085039 4.225 2.40e-05 ***
ns(NUMERIC_AGE, df = s)2 0.230563 0.099705 2.312 0.02076 *
ns(NUMERIC_AGE, df = s)3 -0.369434 0.067707 -5.456 4.92e-08 ***
ns(NUMERIC_AGE, df = s)4 0.450814 0.209545 2.151 0.03146 *
ns(NUMERIC_AGE, df = s)5 -0.482863 0.104079 -4.639 3.52e-06 ***
PM_VISIT_LAST_2_YRS 0.302190 0.028790 10.496 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.279258 0.028960 9.643 < 2e-16 ***
AF_25K_GIFT 0.588082 0.043198 13.614 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.259590 0.006763 38.383 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.913908 0.042293 68.898 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.557706 0.025538 21.839 < 2e-16 ***
MG_250K_PLUS 1.166991 0.070045 16.661 < 2e-16 ***
Alumnus -0.568739 0.021427 -26.543 < 2e-16 ***
SEASON_TICKET_YEARS -0.018724 0.004505 -4.156 3.25e-05 ***
AFFINITY_SCORE 0.158592 0.006794 23.343 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.123273 0.022106 -5.576 2.49e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.208271 0.028348 7.347 2.11e-13 ***
MG_PR_MODEL_DESCTop Tier 0.600171 0.028343 21.175 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.961 on 18759 degrees of freedom
Multiple R-squared: 0.7273, Adjusted R-squared: 0.7271
F-statistic: 2780 on 18 and 18759 DF, p-value: < 2.2e-16
[[5]][[3]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3972 -0.6247 -0.1861 0.4948 6.3062
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.331319 0.095104 3.484 0.000496 ***
ns(NUMERIC_AGE, df = s)1 0.310424 0.086897 3.572 0.000355 ***
ns(NUMERIC_AGE, df = s)2 0.161271 0.101850 1.583 0.113343
ns(NUMERIC_AGE, df = s)3 -0.422841 0.069459 -6.088 1.17e-09 ***
ns(NUMERIC_AGE, df = s)4 0.320024 0.215424 1.486 0.137415
ns(NUMERIC_AGE, df = s)5 -0.497399 0.113736 -4.373 1.23e-05 ***
PM_VISIT_LAST_2_YRS 0.319925 0.028793 11.111 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.263066 0.028893 9.105 < 2e-16 ***
AF_25K_GIFT 0.613941 0.042757 14.359 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.262874 0.006788 38.726 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.916592 0.042389 68.805 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.563381 0.025631 21.980 < 2e-16 ***
MG_250K_PLUS 1.089595 0.071116 15.321 < 2e-16 ***
Alumnus -0.574934 0.021439 -26.818 < 2e-16 ***
SEASON_TICKET_YEARS -0.020092 0.004420 -4.546 5.50e-06 ***
AFFINITY_SCORE 0.160383 0.006823 23.506 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.110457 0.022177 -4.981 6.39e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.207385 0.028463 7.286 3.32e-13 ***
MG_PR_MODEL_DESCTop Tier 0.594950 0.028351 20.985 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.964 on 18759 degrees of freedom
Multiple R-squared: 0.726, Adjusted R-squared: 0.7257
F-statistic: 2761 on 18 and 18759 DF, p-value: < 2.2e-16
[[5]][[4]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4015 -0.6252 -0.1848 0.4979 6.2642
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.274694 0.093894 2.926 0.003442 **
ns(NUMERIC_AGE, df = s)1 0.375171 0.085767 4.374 1.22e-05 ***
ns(NUMERIC_AGE, df = s)2 0.250632 0.100418 2.496 0.012573 *
ns(NUMERIC_AGE, df = s)3 -0.405832 0.069359 -5.851 4.96e-09 ***
ns(NUMERIC_AGE, df = s)4 0.526632 0.212578 2.477 0.013245 *
ns(NUMERIC_AGE, df = s)5 -0.418507 0.114422 -3.658 0.000255 ***
PM_VISIT_LAST_2_YRS 0.319594 0.028956 11.037 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.287850 0.028926 9.951 < 2e-16 ***
AF_25K_GIFT 0.625386 0.042895 14.579 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.259398 0.006796 38.169 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.925863 0.042344 69.098 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.578451 0.025528 22.659 < 2e-16 ***
MG_250K_PLUS 1.092788 0.071386 15.308 < 2e-16 ***
Alumnus -0.572718 0.021470 -26.675 < 2e-16 ***
SEASON_TICKET_YEARS -0.019652 0.004442 -4.424 9.76e-06 ***
AFFINITY_SCORE 0.155681 0.006829 22.797 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.124106 0.022225 -5.584 2.38e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.192985 0.028538 6.762 1.40e-11 ***
MG_PR_MODEL_DESCTop Tier 0.576692 0.028387 20.315 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9648 on 18759 degrees of freedom
Multiple R-squared: 0.7244, Adjusted R-squared: 0.7242
F-statistic: 2740 on 18 and 18759 DF, p-value: < 2.2e-16
[[5]][[5]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.1069 -0.6275 -0.1859 0.4998 6.2731
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.304096 0.094649 3.213 0.00132 **
ns(NUMERIC_AGE, df = s)1 0.349770 0.086479 4.045 5.26e-05 ***
ns(NUMERIC_AGE, df = s)2 0.205294 0.101354 2.026 0.04283 *
ns(NUMERIC_AGE, df = s)3 -0.385610 0.069824 -5.523 3.38e-08 ***
ns(NUMERIC_AGE, df = s)4 0.411601 0.214714 1.917 0.05526 .
ns(NUMERIC_AGE, df = s)5 -0.467467 0.116063 -4.028 5.65e-05 ***
PM_VISIT_LAST_2_YRS 0.339281 0.029088 11.664 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.276994 0.029135 9.507 < 2e-16 ***
AF_25K_GIFT 0.607389 0.043052 14.108 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.258651 0.006802 38.026 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.917820 0.042572 68.539 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.559188 0.025716 21.745 < 2e-16 ***
MG_250K_PLUS 1.097125 0.071871 15.265 < 2e-16 ***
Alumnus -0.573888 0.021552 -26.628 < 2e-16 ***
SEASON_TICKET_YEARS -0.020764 0.004515 -4.599 4.26e-06 ***
AFFINITY_SCORE 0.158616 0.006845 23.171 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.116487 0.022304 -5.223 1.78e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.221006 0.028528 7.747 9.88e-15 ***
MG_PR_MODEL_DESCTop Tier 0.594491 0.028418 20.920 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9682 on 18759 degrees of freedom
Multiple R-squared: 0.723, Adjusted R-squared: 0.7227
F-statistic: 2720 on 18 and 18759 DF, p-value: < 2.2e-16
[[5]][[6]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4361 -0.6200 -0.1867 0.4945 6.2946
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.278270 0.093230 2.985 0.00284 **
ns(NUMERIC_AGE, df = s)1 0.410210 0.085208 4.814 1.49e-06 ***
ns(NUMERIC_AGE, df = s)2 0.233923 0.099899 2.342 0.01921 *
ns(NUMERIC_AGE, df = s)3 -0.361744 0.069187 -5.229 1.73e-07 ***
ns(NUMERIC_AGE, df = s)4 0.484563 0.211617 2.290 0.02204 *
ns(NUMERIC_AGE, df = s)5 -0.497700 0.114815 -4.335 1.47e-05 ***
PM_VISIT_LAST_2_YRS 0.283623 0.029129 9.737 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.300906 0.028885 10.417 < 2e-16 ***
AF_25K_GIFT 0.614608 0.043478 14.136 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.260185 0.006788 38.331 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.913073 0.042421 68.670 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.562183 0.025537 22.014 < 2e-16 ***
MG_250K_PLUS 1.139511 0.072230 15.776 < 2e-16 ***
Alumnus -0.581322 0.021355 -27.222 < 2e-16 ***
SEASON_TICKET_YEARS -0.018984 0.004442 -4.274 1.93e-05 ***
AFFINITY_SCORE 0.158698 0.006821 23.265 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.132689 0.022198 -5.978 2.31e-09 ***
MG_PR_MODEL_DESCMiddle Tier 0.191770 0.028444 6.742 1.61e-11 ***
MG_PR_MODEL_DESCTop Tier 0.565934 0.028285 20.008 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9624 on 18759 degrees of freedom
Multiple R-squared: 0.7255, Adjusted R-squared: 0.7252
F-statistic: 2754 on 18 and 18759 DF, p-value: < 2.2e-16
[[5]][[7]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3994 -0.6266 -0.1874 0.4958 6.2636
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.283086 0.093027 3.043 0.002345 **
ns(NUMERIC_AGE, df = s)1 0.349749 0.084895 4.120 3.81e-05 ***
ns(NUMERIC_AGE, df = s)2 0.214313 0.099606 2.152 0.031441 *
ns(NUMERIC_AGE, df = s)3 -0.438830 0.069369 -6.326 2.57e-10 ***
ns(NUMERIC_AGE, df = s)4 0.456633 0.211028 2.164 0.030489 *
ns(NUMERIC_AGE, df = s)5 -0.411888 0.116315 -3.541 0.000399 ***
PM_VISIT_LAST_2_YRS 0.305546 0.028936 10.560 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.299356 0.028897 10.359 < 2e-16 ***
AF_25K_GIFT 0.623798 0.042930 14.531 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.263053 0.006784 38.776 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.899286 0.042408 68.366 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.567406 0.025584 22.179 < 2e-16 ***
MG_250K_PLUS 1.092580 0.072594 15.051 < 2e-16 ***
Alumnus -0.570026 0.021460 -26.562 < 2e-16 ***
SEASON_TICKET_YEARS -0.018948 0.004532 -4.181 2.92e-05 ***
AFFINITY_SCORE 0.157965 0.006824 23.149 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.109675 0.022215 -4.937 8.01e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.212359 0.028477 7.457 9.23e-14 ***
MG_PR_MODEL_DESCTop Tier 0.578368 0.028332 20.414 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9637 on 18759 degrees of freedom
Multiple R-squared: 0.7248, Adjusted R-squared: 0.7245
F-statistic: 2745 on 18 and 18759 DF, p-value: < 2.2e-16
[[5]][[8]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4493 -0.6252 -0.1804 0.4957 6.2555
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.242819 0.094508 2.569 0.010198 *
ns(NUMERIC_AGE, df = s)1 0.400320 0.086354 4.636 3.58e-06 ***
ns(NUMERIC_AGE, df = s)2 0.280377 0.101216 2.770 0.005610 **
ns(NUMERIC_AGE, df = s)3 -0.386612 0.069558 -5.558 2.76e-08 ***
ns(NUMERIC_AGE, df = s)4 0.554027 0.214110 2.588 0.009673 **
ns(NUMERIC_AGE, df = s)5 -0.417324 0.114021 -3.660 0.000253 ***
PM_VISIT_LAST_2_YRS 0.318603 0.029185 10.917 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.287104 0.029101 9.866 < 2e-16 ***
AF_25K_GIFT 0.576878 0.043685 13.206 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.258682 0.006836 37.844 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.901273 0.042488 68.284 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.555337 0.025751 21.565 < 2e-16 ***
MG_250K_PLUS 1.127600 0.072302 15.596 < 2e-16 ***
Alumnus -0.568027 0.021564 -26.341 < 2e-16 ***
SEASON_TICKET_YEARS -0.018766 0.004485 -4.185 2.87e-05 ***
AFFINITY_SCORE 0.160057 0.006853 23.356 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.117462 0.022260 -5.277 1.33e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.200356 0.028555 7.016 2.35e-12 ***
MG_PR_MODEL_DESCTop Tier 0.585297 0.028422 20.593 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9682 on 18759 degrees of freedom
Multiple R-squared: 0.7218, Adjusted R-squared: 0.7216
F-statistic: 2704 on 18 and 18759 DF, p-value: < 2.2e-16
[[5]][[9]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3846 -0.6212 -0.1874 0.4930 6.2869
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.240996 0.093787 2.570 0.0102 *
ns(NUMERIC_AGE, df = s)1 0.398169 0.085670 4.648 3.38e-06 ***
ns(NUMERIC_AGE, df = s)2 0.265551 0.100441 2.644 0.0082 **
ns(NUMERIC_AGE, df = s)3 -0.358989 0.069381 -5.174 2.31e-07 ***
ns(NUMERIC_AGE, df = s)4 0.521546 0.212440 2.455 0.0141 *
ns(NUMERIC_AGE, df = s)5 -0.457178 0.115156 -3.970 7.21e-05 ***
PM_VISIT_LAST_2_YRS 0.324927 0.028819 11.275 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.276194 0.028930 9.547 < 2e-16 ***
AF_25K_GIFT 0.569050 0.043065 13.214 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.259652 0.006808 38.138 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.914343 0.042620 68.379 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.576876 0.025602 22.532 < 2e-16 ***
MG_250K_PLUS 1.071882 0.071627 14.965 < 2e-16 ***
Alumnus -0.570608 0.021474 -26.572 < 2e-16 ***
SEASON_TICKET_YEARS -0.022116 0.004461 -4.958 7.19e-07 ***
AFFINITY_SCORE 0.159975 0.006819 23.460 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.119662 0.022296 -5.367 8.10e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.208362 0.028645 7.274 3.63e-13 ***
MG_PR_MODEL_DESCTop Tier 0.588830 0.028309 20.800 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9655 on 18759 degrees of freedom
Multiple R-squared: 0.7255, Adjusted R-squared: 0.7252
F-statistic: 2754 on 18 and 18759 DF, p-value: < 2.2e-16
[[5]][[10]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3449 -0.6302 -0.1870 0.5008 6.2783
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.245968 0.093356 2.635 0.00843 **
ns(NUMERIC_AGE, df = s)1 0.385387 0.085217 4.522 6.15e-06 ***
ns(NUMERIC_AGE, df = s)2 0.270300 0.099870 2.707 0.00681 **
ns(NUMERIC_AGE, df = s)3 -0.364235 0.069246 -5.260 1.46e-07 ***
ns(NUMERIC_AGE, df = s)4 0.538473 0.211455 2.547 0.01089 *
ns(NUMERIC_AGE, df = s)5 -0.465707 0.114036 -4.084 4.45e-05 ***
PM_VISIT_LAST_2_YRS 0.311610 0.028905 10.781 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.283431 0.029058 9.754 < 2e-16 ***
AF_25K_GIFT 0.612449 0.042930 14.266 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.254815 0.006813 37.400 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.919716 0.042574 68.580 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.582585 0.025656 22.708 < 2e-16 ***
MG_250K_PLUS 1.062557 0.071953 14.767 < 2e-16 ***
Alumnus -0.567552 0.021520 -26.373 < 2e-16 ***
SEASON_TICKET_YEARS -0.021359 0.004514 -4.732 2.24e-06 ***
AFFINITY_SCORE 0.161414 0.006832 23.628 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.112613 0.022355 -5.037 4.76e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.203528 0.028624 7.111 1.20e-12 ***
MG_PR_MODEL_DESCTop Tier 0.577537 0.028417 20.323 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9675 on 18755 degrees of freedom
Multiple R-squared: 0.7231, Adjusted R-squared: 0.7229
F-statistic: 2721 on 18 and 18755 DF, p-value: < 2.2e-16
[[6]]
[[6]][[1]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4417 -0.6241 -0.1817 0.4929 5.4637
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.161509 0.102127 1.581 0.11379
ns(NUMERIC_AGE, df = s)1 0.449311 0.090479 4.966 6.90e-07 ***
ns(NUMERIC_AGE, df = s)2 0.445612 0.102831 4.333 1.48e-05 ***
ns(NUMERIC_AGE, df = s)3 0.244518 0.103602 2.360 0.01828 *
ns(NUMERIC_AGE, df = s)4 -0.227344 0.081194 -2.800 0.00512 **
ns(NUMERIC_AGE, df = s)5 0.623163 0.226471 2.752 0.00594 **
ns(NUMERIC_AGE, df = s)6 -0.627560 0.124148 -5.055 4.35e-07 ***
PM_VISIT_LAST_2_YRS 0.311088 0.028731 10.827 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.275203 0.028876 9.530 < 2e-16 ***
AF_25K_GIFT 0.616886 0.042562 14.494 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.261364 0.006781 38.544 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.924384 0.042341 69.067 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.555818 0.025512 21.787 < 2e-16 ***
MG_250K_PLUS 1.125455 0.071597 15.719 < 2e-16 ***
Alumnus -0.563399 0.021700 -25.963 < 2e-16 ***
SEASON_TICKET_YEARS -0.019191 0.004461 -4.302 1.70e-05 ***
AFFINITY_SCORE 0.159102 0.006806 23.378 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.117204 0.022126 -5.297 1.19e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.214947 0.028347 7.583 3.54e-14 ***
MG_PR_MODEL_DESCTop Tier 0.596042 0.028194 21.141 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9609 on 18758 degrees of freedom
Multiple R-squared: 0.7272, Adjusted R-squared: 0.7269
F-statistic: 2631 on 19 and 18758 DF, p-value: < 2.2e-16
[[6]][[2]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4882 -0.6226 -0.1777 0.4910 6.3082
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.140629 0.102735 1.369 0.17106
ns(NUMERIC_AGE, df = s)1 0.453043 0.091284 4.963 7.00e-07 ***
ns(NUMERIC_AGE, df = s)2 0.495750 0.104207 4.757 1.98e-06 ***
ns(NUMERIC_AGE, df = s)3 0.248431 0.103865 2.392 0.01677 *
ns(NUMERIC_AGE, df = s)4 -0.210035 0.079848 -2.630 0.00853 **
ns(NUMERIC_AGE, df = s)5 0.717798 0.226054 3.175 0.00150 **
ns(NUMERIC_AGE, df = s)6 -0.564545 0.112837 -5.003 5.69e-07 ***
PM_VISIT_LAST_2_YRS 0.301663 0.028782 10.481 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.280519 0.028954 9.688 < 2e-16 ***
AF_25K_GIFT 0.589883 0.043189 13.658 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.260352 0.006766 38.478 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.907212 0.042331 68.678 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.554858 0.025543 21.722 < 2e-16 ***
MG_250K_PLUS 1.166120 0.070024 16.653 < 2e-16 ***
Alumnus -0.557103 0.021692 -25.682 < 2e-16 ***
SEASON_TICKET_YEARS -0.018645 0.004504 -4.140 3.49e-05 ***
AFFINITY_SCORE 0.159680 0.006798 23.488 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.122390 0.022102 -5.538 3.11e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.208786 0.028342 7.367 1.82e-13 ***
MG_PR_MODEL_DESCTop Tier 0.601227 0.028338 21.216 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9607 on 18758 degrees of freedom
Multiple R-squared: 0.7275, Adjusted R-squared: 0.7272
F-statistic: 2636 on 19 and 18758 DF, p-value: < 2.2e-16
[[6]][[3]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4112 -0.6254 -0.1831 0.4952 6.3333
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.205039 0.105406 1.945 0.051763 .
ns(NUMERIC_AGE, df = s)1 0.399056 0.093598 4.264 2.02e-05 ***
ns(NUMERIC_AGE, df = s)2 0.415436 0.106898 3.886 0.000102 ***
ns(NUMERIC_AGE, df = s)3 0.176561 0.106375 1.660 0.096972 .
ns(NUMERIC_AGE, df = s)4 -0.278527 0.082249 -3.386 0.000710 ***
ns(NUMERIC_AGE, df = s)5 0.550571 0.232668 2.366 0.017975 *
ns(NUMERIC_AGE, df = s)6 -0.577993 0.123692 -4.673 2.99e-06 ***
PM_VISIT_LAST_2_YRS 0.319663 0.028788 11.104 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.263825 0.028888 9.133 < 2e-16 ***
AF_25K_GIFT 0.615479 0.042751 14.397 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.263509 0.006792 38.798 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.910252 0.042439 68.575 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.560671 0.025643 21.865 < 2e-16 ***
MG_250K_PLUS 1.089131 0.071102 15.318 < 2e-16 ***
Alumnus -0.565002 0.021702 -26.034 < 2e-16 ***
SEASON_TICKET_YEARS -0.020049 0.004419 -4.537 5.74e-06 ***
AFFINITY_SCORE 0.161395 0.006829 23.633 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.109735 0.022174 -4.949 7.53e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.207843 0.028459 7.303 2.92e-13 ***
MG_PR_MODEL_DESCTop Tier 0.596083 0.028348 21.027 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9639 on 18758 degrees of freedom
Multiple R-squared: 0.7261, Adjusted R-squared: 0.7258
F-statistic: 2617 on 19 and 18758 DF, p-value: < 2.2e-16
[[6]][[4]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4164 -0.6254 -0.1826 0.4979 6.2955
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.155837 0.102731 1.517 0.12930
ns(NUMERIC_AGE, df = s)1 0.455519 0.090985 5.007 5.59e-07 ***
ns(NUMERIC_AGE, df = s)2 0.469975 0.103308 4.549 5.42e-06 ***
ns(NUMERIC_AGE, df = s)3 0.242309 0.104399 2.321 0.02030 *
ns(NUMERIC_AGE, df = s)4 -0.248052 0.082820 -2.995 0.00275 **
ns(NUMERIC_AGE, df = s)5 0.736255 0.227884 3.231 0.00124 **
ns(NUMERIC_AGE, df = s)6 -0.522828 0.126773 -4.124 3.74e-05 ***
PM_VISIT_LAST_2_YRS 0.318796 0.028951 11.011 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.288959 0.028923 9.991 < 2e-16 ***
AF_25K_GIFT 0.626774 0.042888 14.614 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.260093 0.006800 38.246 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.919263 0.042401 68.848 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.575730 0.025540 22.542 < 2e-16 ***
MG_250K_PLUS 1.092521 0.071370 15.308 < 2e-16 ***
Alumnus -0.562627 0.021762 -25.854 < 2e-16 ***
SEASON_TICKET_YEARS -0.019566 0.004442 -4.405 1.06e-05 ***
AFFINITY_SCORE 0.156739 0.006837 22.926 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.123505 0.022222 -5.558 2.77e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.193407 0.028533 6.778 1.25e-11 ***
MG_PR_MODEL_DESCTop Tier 0.577673 0.028384 20.352 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9646 on 18758 degrees of freedom
Multiple R-squared: 0.7246, Adjusted R-squared: 0.7243
F-statistic: 2597 on 19 and 18758 DF, p-value: < 2.2e-16
[[6]][[5]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.1155 -0.6278 -0.1842 0.4990 6.3037
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.188382 0.103797 1.815 0.06955 .
ns(NUMERIC_AGE, df = s)1 0.424115 0.091864 4.617 3.92e-06 ***
ns(NUMERIC_AGE, df = s)2 0.432669 0.104605 4.136 3.55e-05 ***
ns(NUMERIC_AGE, df = s)3 0.206241 0.105209 1.960 0.04998 *
ns(NUMERIC_AGE, df = s)4 -0.240483 0.082253 -2.924 0.00346 **
ns(NUMERIC_AGE, df = s)5 0.616767 0.230280 2.678 0.00741 **
ns(NUMERIC_AGE, df = s)6 -0.564864 0.126432 -4.468 7.95e-06 ***
PM_VISIT_LAST_2_YRS 0.338502 0.029083 11.639 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.278173 0.029132 9.549 < 2e-16 ***
AF_25K_GIFT 0.608798 0.043045 14.143 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.259287 0.006806 38.099 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.911506 0.042624 68.307 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.556232 0.025731 21.617 < 2e-16 ***
MG_250K_PLUS 1.096862 0.071855 15.265 < 2e-16 ***
Alumnus -0.563400 0.021863 -25.769 < 2e-16 ***
SEASON_TICKET_YEARS -0.020632 0.004514 -4.571 4.89e-06 ***
AFFINITY_SCORE 0.159678 0.006853 23.300 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.115959 0.022300 -5.200 2.02e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.221331 0.028523 7.760 8.95e-15 ***
MG_PR_MODEL_DESCTop Tier 0.595406 0.028414 20.954 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.968 on 18758 degrees of freedom
Multiple R-squared: 0.7231, Adjusted R-squared: 0.7228
F-statistic: 2578 on 19 and 18758 DF, p-value: < 2.2e-16
[[6]][[6]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4513 -0.6222 -0.1835 0.4958 6.3237
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.165886 0.102120 1.624 0.10430
ns(NUMERIC_AGE, df = s)1 0.484415 0.090411 5.358 8.52e-08 ***
ns(NUMERIC_AGE, df = s)2 0.472897 0.102911 4.595 4.35e-06 ***
ns(NUMERIC_AGE, df = s)3 0.236019 0.103611 2.278 0.02274 *
ns(NUMERIC_AGE, df = s)4 -0.221179 0.081433 -2.716 0.00661 **
ns(NUMERIC_AGE, df = s)5 0.678408 0.226636 2.993 0.00276 **
ns(NUMERIC_AGE, df = s)6 -0.588336 0.125270 -4.697 2.66e-06 ***
PM_VISIT_LAST_2_YRS 0.283328 0.029123 9.729 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.302050 0.028883 10.458 < 2e-16 ***
AF_25K_GIFT 0.615960 0.043472 14.169 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.260861 0.006792 38.405 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.907219 0.042468 68.457 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.559504 0.025550 21.898 < 2e-16 ***
MG_250K_PLUS 1.139550 0.072215 15.780 < 2e-16 ***
Alumnus -0.571090 0.021664 -26.361 < 2e-16 ***
SEASON_TICKET_YEARS -0.018906 0.004441 -4.257 2.08e-05 ***
AFFINITY_SCORE 0.159629 0.006827 23.382 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.131877 0.022196 -5.942 2.87e-09 ***
MG_PR_MODEL_DESCMiddle Tier 0.192339 0.028440 6.763 1.39e-11 ***
MG_PR_MODEL_DESCTop Tier 0.567138 0.028285 20.051 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9622 on 18758 degrees of freedom
Multiple R-squared: 0.7256, Adjusted R-squared: 0.7253
F-statistic: 2611 on 19 and 18758 DF, p-value: < 2.2e-16
[[6]][[7]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4131 -0.6255 -0.1844 0.4966 6.2897
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.165449 0.102705 1.611 0.107215
ns(NUMERIC_AGE, df = s)1 0.438551 0.091123 4.813 1.50e-06 ***
ns(NUMERIC_AGE, df = s)2 0.448963 0.104016 4.316 1.59e-05 ***
ns(NUMERIC_AGE, df = s)3 0.225724 0.103765 2.175 0.029617 *
ns(NUMERIC_AGE, df = s)4 -0.300647 0.081916 -3.670 0.000243 ***
ns(NUMERIC_AGE, df = s)5 0.667750 0.226961 2.942 0.003263 **
ns(NUMERIC_AGE, df = s)6 -0.484865 0.127072 -3.816 0.000136 ***
PM_VISIT_LAST_2_YRS 0.305441 0.028931 10.558 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.300279 0.028894 10.392 < 2e-16 ***
AF_25K_GIFT 0.624477 0.042922 14.549 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.263622 0.006788 38.836 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.893931 0.042455 68.165 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.564849 0.025598 22.066 < 2e-16 ***
MG_250K_PLUS 1.092780 0.072582 15.056 < 2e-16 ***
Alumnus -0.561241 0.021711 -25.851 < 2e-16 ***
SEASON_TICKET_YEARS -0.018881 0.004532 -4.166 3.11e-05 ***
AFFINITY_SCORE 0.158836 0.006829 23.258 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.108647 0.022214 -4.891 1.01e-06 ***
MG_PR_MODEL_DESCMiddle Tier 0.213204 0.028475 7.488 7.33e-14 ***
MG_PR_MODEL_DESCTop Tier 0.579905 0.028333 20.467 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9635 on 18758 degrees of freedom
Multiple R-squared: 0.7249, Adjusted R-squared: 0.7246
F-statistic: 2602 on 19 and 18758 DF, p-value: < 2.2e-16
[[6]][[8]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4624 -0.6277 -0.1807 0.4960 6.2873
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.117734 0.104527 1.126 0.260034
ns(NUMERIC_AGE, df = s)1 0.492262 0.092844 5.302 1.16e-07 ***
ns(NUMERIC_AGE, df = s)2 0.517367 0.105802 4.890 1.02e-06 ***
ns(NUMERIC_AGE, df = s)3 0.278630 0.105841 2.633 0.008482 **
ns(NUMERIC_AGE, df = s)4 -0.221484 0.083447 -2.654 0.007957 **
ns(NUMERIC_AGE, df = s)5 0.769634 0.231079 3.331 0.000868 ***
ns(NUMERIC_AGE, df = s)6 -0.516069 0.126160 -4.091 4.32e-05 ***
PM_VISIT_LAST_2_YRS 0.318214 0.029180 10.905 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.287937 0.029096 9.896 < 2e-16 ***
AF_25K_GIFT 0.578544 0.043678 13.246 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.259358 0.006840 37.920 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.895109 0.042535 68.063 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.552872 0.025760 21.462 < 2e-16 ***
MG_250K_PLUS 1.126702 0.072287 15.587 < 2e-16 ***
Alumnus -0.558070 0.021841 -25.552 < 2e-16 ***
SEASON_TICKET_YEARS -0.018726 0.004484 -4.177 2.97e-05 ***
AFFINITY_SCORE 0.160958 0.006858 23.471 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.116829 0.022257 -5.249 1.55e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.200768 0.028551 7.032 2.11e-12 ***
MG_PR_MODEL_DESCTop Tier 0.586164 0.028419 20.626 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.968 on 18758 degrees of freedom
Multiple R-squared: 0.722, Adjusted R-squared: 0.7217
F-statistic: 2564 on 19 and 18758 DF, p-value: < 2.2e-16
[[6]][[9]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4004 -0.6227 -0.1836 0.4913 6.3203
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.108097 0.103590 1.044 0.296725
ns(NUMERIC_AGE, df = s)1 0.488379 0.091969 5.310 1.11e-07 ***
ns(NUMERIC_AGE, df = s)2 0.522123 0.104958 4.975 6.60e-07 ***
ns(NUMERIC_AGE, df = s)3 0.275644 0.104661 2.634 0.008454 **
ns(NUMERIC_AGE, df = s)4 -0.195021 0.081942 -2.380 0.017324 *
ns(NUMERIC_AGE, df = s)5 0.755605 0.228699 3.304 0.000955 ***
ns(NUMERIC_AGE, df = s)6 -0.556749 0.125394 -4.440 9.05e-06 ***
PM_VISIT_LAST_2_YRS 0.324496 0.028812 11.262 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.277191 0.028923 9.584 < 2e-16 ***
AF_25K_GIFT 0.570775 0.043057 13.256 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.260392 0.006812 38.227 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.907935 0.042658 68.168 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.574039 0.025610 22.414 < 2e-16 ***
MG_250K_PLUS 1.072012 0.071608 14.971 < 2e-16 ***
Alumnus -0.559832 0.021726 -25.767 < 2e-16 ***
SEASON_TICKET_YEARS -0.022048 0.004460 -4.944 7.73e-07 ***
AFFINITY_SCORE 0.160985 0.006823 23.593 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.118475 0.022293 -5.314 1.08e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.208956 0.028640 7.296 3.08e-13 ***
MG_PR_MODEL_DESCTop Tier 0.590000 0.028306 20.844 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9652 on 18758 degrees of freedom
Multiple R-squared: 0.7256, Adjusted R-squared: 0.7253
F-statistic: 2611 on 19 and 18758 DF, p-value: < 2.2e-16
[[6]][[10]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3596 -0.6299 -0.1862 0.4999 6.3119
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.110230 0.103064 1.070 0.284844
ns(NUMERIC_AGE, df = s)1 0.481905 0.091466 5.269 1.39e-07 ***
ns(NUMERIC_AGE, df = s)2 0.516898 0.104203 4.960 7.09e-07 ***
ns(NUMERIC_AGE, df = s)3 0.277045 0.104323 2.656 0.007922 **
ns(NUMERIC_AGE, df = s)4 -0.193612 0.082869 -2.336 0.019483 *
ns(NUMERIC_AGE, df = s)5 0.776015 0.227918 3.405 0.000664 ***
ns(NUMERIC_AGE, df = s)6 -0.572936 0.126296 -4.536 5.76e-06 ***
PM_VISIT_LAST_2_YRS 0.310653 0.028899 10.750 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.284572 0.029052 9.795 < 2e-16 ***
AF_25K_GIFT 0.614159 0.042922 14.309 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.255594 0.006818 37.491 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.912890 0.042624 68.340 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.579588 0.025667 22.581 < 2e-16 ***
MG_250K_PLUS 1.061861 0.071934 14.762 < 2e-16 ***
Alumnus -0.556824 0.021789 -25.556 < 2e-16 ***
SEASON_TICKET_YEARS -0.021328 0.004513 -4.726 2.31e-06 ***
AFFINITY_SCORE 0.162351 0.006835 23.752 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.111699 0.022351 -4.997 5.86e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.204208 0.028618 7.136 9.99e-13 ***
MG_PR_MODEL_DESCTop Tier 0.578891 0.028414 20.373 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9673 on 18754 degrees of freedom
Multiple R-squared: 0.7233, Adjusted R-squared: 0.723
F-statistic: 2580 on 19 and 18754 DF, p-value: < 2.2e-16
[[7]]
[[7]][[1]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4413 -0.6242 -0.1800 0.4921 5.4641
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.102258 0.108266 0.945 0.34492
ns(NUMERIC_AGE, df = s)1 0.510971 0.095517 5.350 8.92e-08 ***
ns(NUMERIC_AGE, df = s)2 0.544020 0.120432 4.517 6.30e-06 ***
ns(NUMERIC_AGE, df = s)3 0.425155 0.106077 4.008 6.15e-05 ***
ns(NUMERIC_AGE, df = s)4 0.271595 0.114011 2.382 0.01722 *
ns(NUMERIC_AGE, df = s)5 -0.180979 0.084863 -2.133 0.03297 *
ns(NUMERIC_AGE, df = s)6 0.723239 0.242227 2.986 0.00283 **
ns(NUMERIC_AGE, df = s)7 -0.636151 0.132839 -4.789 1.69e-06 ***
PM_VISIT_LAST_2_YRS 0.311329 0.028730 10.836 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.275127 0.028874 9.528 < 2e-16 ***
AF_25K_GIFT 0.616899 0.042558 14.495 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.261342 0.006780 38.543 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.924686 0.042339 69.077 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.556075 0.025513 21.796 < 2e-16 ***
MG_250K_PLUS 1.125042 0.071603 15.712 < 2e-16 ***
Alumnus -0.563446 0.021681 -25.988 < 2e-16 ***
SEASON_TICKET_YEARS -0.019252 0.004460 -4.316 1.59e-05 ***
AFFINITY_SCORE 0.159029 0.006805 23.371 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.116947 0.022125 -5.286 1.27e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.215291 0.028347 7.595 3.23e-14 ***
MG_PR_MODEL_DESCTop Tier 0.596571 0.028197 21.157 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9608 on 18757 degrees of freedom
Multiple R-squared: 0.7272, Adjusted R-squared: 0.7269
F-statistic: 2500 on 20 and 18757 DF, p-value: < 2.2e-16
[[7]][[2]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4856 -0.6240 -0.1788 0.4903 6.3129
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.096716 0.106796 0.906 0.365152
ns(NUMERIC_AGE, df = s)1 0.491068 0.094256 5.210 1.91e-07 ***
ns(NUMERIC_AGE, df = s)2 0.575352 0.119142 4.829 1.38e-06 ***
ns(NUMERIC_AGE, df = s)3 0.426741 0.104718 4.075 4.62e-05 ***
ns(NUMERIC_AGE, df = s)4 0.272398 0.112845 2.414 0.015792 *
ns(NUMERIC_AGE, df = s)5 -0.185622 0.082800 -2.242 0.024984 *
ns(NUMERIC_AGE, df = s)6 0.805318 0.237965 3.384 0.000715 ***
ns(NUMERIC_AGE, df = s)7 -0.565719 0.120186 -4.707 2.53e-06 ***
PM_VISIT_LAST_2_YRS 0.301844 0.028780 10.488 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.280407 0.028953 9.685 < 2e-16 ***
AF_25K_GIFT 0.589746 0.043186 13.656 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.260258 0.006767 38.461 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.908368 0.042344 68.684 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.555500 0.025551 21.741 < 2e-16 ***
MG_250K_PLUS 1.165449 0.070030 16.642 < 2e-16 ***
Alumnus -0.558534 0.021734 -25.699 < 2e-16 ***
SEASON_TICKET_YEARS -0.018690 0.004504 -4.150 3.34e-05 ***
AFFINITY_SCORE 0.159462 0.006800 23.449 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.122238 0.022101 -5.531 3.23e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.209046 0.028343 7.376 1.70e-13 ***
MG_PR_MODEL_DESCTop Tier 0.601631 0.028341 21.229 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9607 on 18757 degrees of freedom
Multiple R-squared: 0.7276, Adjusted R-squared: 0.7273
F-statistic: 2505 on 20 and 18757 DF, p-value: < 2.2e-16
[[7]][[3]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4090 -0.6257 -0.1826 0.4936 6.3371
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.161157 0.109801 1.468 0.14220
ns(NUMERIC_AGE, df = s)1 0.435524 0.096783 4.500 6.84e-06 ***
ns(NUMERIC_AGE, df = s)2 0.499929 0.122415 4.084 4.45e-05 ***
ns(NUMERIC_AGE, df = s)3 0.347658 0.107615 3.231 0.00124 **
ns(NUMERIC_AGE, df = s)4 0.199849 0.115648 1.728 0.08399 .
ns(NUMERIC_AGE, df = s)5 -0.254522 0.085226 -2.986 0.00283 **
ns(NUMERIC_AGE, df = s)6 0.640322 0.245682 2.606 0.00916 **
ns(NUMERIC_AGE, df = s)7 -0.578993 0.132246 -4.378 1.20e-05 ***
PM_VISIT_LAST_2_YRS 0.319923 0.028786 11.114 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.263739 0.028887 9.130 < 2e-16 ***
AF_25K_GIFT 0.615302 0.042748 14.394 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.263402 0.006793 38.778 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.911190 0.042451 68.577 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.561260 0.025648 21.883 < 2e-16 ***
MG_250K_PLUS 1.088688 0.071105 15.311 < 2e-16 ***
Alumnus -0.566247 0.021732 -26.056 < 2e-16 ***
SEASON_TICKET_YEARS -0.020093 0.004419 -4.547 5.47e-06 ***
AFFINITY_SCORE 0.161209 0.006831 23.601 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.109648 0.022173 -4.945 7.67e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.208058 0.028458 7.311 2.76e-13 ***
MG_PR_MODEL_DESCTop Tier 0.596277 0.028347 21.035 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9638 on 18757 degrees of freedom
Multiple R-squared: 0.7261, Adjusted R-squared: 0.7258
F-statistic: 2487 on 20 and 18757 DF, p-value: < 2.2e-16
[[7]][[4]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4136 -0.6247 -0.1810 0.4969 6.2964
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.095146 0.107842 0.882 0.377642
ns(NUMERIC_AGE, df = s)1 0.510990 0.095101 5.373 7.83e-08 ***
ns(NUMERIC_AGE, df = s)2 0.584610 0.120110 4.867 1.14e-06 ***
ns(NUMERIC_AGE, df = s)3 0.438237 0.105643 4.148 3.36e-05 ***
ns(NUMERIC_AGE, df = s)4 0.293098 0.113689 2.578 0.009943 **
ns(NUMERIC_AGE, df = s)5 -0.223607 0.084934 -2.633 0.008477 **
ns(NUMERIC_AGE, df = s)6 0.863976 0.241406 3.579 0.000346 ***
ns(NUMERIC_AGE, df = s)7 -0.498940 0.133433 -3.739 0.000185 ***
PM_VISIT_LAST_2_YRS 0.319101 0.028948 11.023 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.288949 0.028921 9.991 < 2e-16 ***
AF_25K_GIFT 0.626451 0.042886 14.607 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.259968 0.006800 38.231 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.920790 0.042410 68.871 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.576489 0.025543 22.569 < 2e-16 ***
MG_250K_PLUS 1.091590 0.071369 15.295 < 2e-16 ***
Alumnus -0.563961 0.021742 -25.939 < 2e-16 ***
SEASON_TICKET_YEARS -0.019633 0.004441 -4.420 9.91e-06 ***
AFFINITY_SCORE 0.156521 0.006837 22.895 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.123333 0.022220 -5.550 2.89e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.193742 0.028532 6.790 1.15e-11 ***
MG_PR_MODEL_DESCTop Tier 0.578168 0.028384 20.370 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9645 on 18757 degrees of freedom
Multiple R-squared: 0.7246, Adjusted R-squared: 0.7243
F-statistic: 2468 on 20 and 18757 DF, p-value: < 2.2e-16
[[7]][[5]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.1156 -0.6280 -0.1823 0.5007 6.3131
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.138251 0.109068 1.268 0.204969
ns(NUMERIC_AGE, df = s)1 0.474572 0.096119 4.937 7.99e-07 ***
ns(NUMERIC_AGE, df = s)2 0.519096 0.121560 4.270 1.96e-05 ***
ns(NUMERIC_AGE, df = s)3 0.396831 0.106902 3.712 0.000206 ***
ns(NUMERIC_AGE, df = s)4 0.222147 0.114929 1.933 0.053262 .
ns(NUMERIC_AGE, df = s)5 -0.193047 0.085581 -2.256 0.024100 *
ns(NUMERIC_AGE, df = s)6 0.702642 0.244537 2.873 0.004066 **
ns(NUMERIC_AGE, df = s)7 -0.583393 0.135193 -4.315 1.60e-05 ***
PM_VISIT_LAST_2_YRS 0.338703 0.029080 11.647 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.278390 0.029130 9.557 < 2e-16 ***
AF_25K_GIFT 0.608615 0.043042 14.140 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.259237 0.006805 38.095 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.911844 0.042623 68.316 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.556515 0.025735 21.625 < 2e-16 ***
MG_250K_PLUS 1.096550 0.071858 15.260 < 2e-16 ***
Alumnus -0.563687 0.021852 -25.795 < 2e-16 ***
SEASON_TICKET_YEARS -0.020677 0.004514 -4.581 4.66e-06 ***
AFFINITY_SCORE 0.159592 0.006853 23.288 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.115784 0.022299 -5.192 2.10e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.221502 0.028524 7.766 8.55e-15 ***
MG_PR_MODEL_DESCTop Tier 0.595680 0.028414 20.964 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9679 on 18757 degrees of freedom
Multiple R-squared: 0.7232, Adjusted R-squared: 0.7229
F-statistic: 2450 on 20 and 18757 DF, p-value: < 2.2e-16
[[7]][[6]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4494 -0.6206 -0.1827 0.4946 6.3287
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.102971 0.107197 0.961 0.336773
ns(NUMERIC_AGE, df = s)1 0.536046 0.094521 5.671 1.44e-08 ***
ns(NUMERIC_AGE, df = s)2 0.600646 0.119553 5.024 5.10e-07 ***
ns(NUMERIC_AGE, df = s)3 0.431191 0.105093 4.103 4.10e-05 ***
ns(NUMERIC_AGE, df = s)4 0.279043 0.113074 2.468 0.013604 *
ns(NUMERIC_AGE, df = s)5 -0.186792 0.084786 -2.203 0.027600 *
ns(NUMERIC_AGE, df = s)6 0.803414 0.240417 3.342 0.000834 ***
ns(NUMERIC_AGE, df = s)7 -0.585442 0.134084 -4.366 1.27e-05 ***
PM_VISIT_LAST_2_YRS 0.283919 0.029122 9.749 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.301950 0.028880 10.455 < 2e-16 ***
AF_25K_GIFT 0.615734 0.043468 14.165 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.260794 0.006791 38.401 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.908517 0.042471 68.483 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.560276 0.025554 21.925 < 2e-16 ***
MG_250K_PLUS 1.138830 0.072215 15.770 < 2e-16 ***
Alumnus -0.572302 0.021646 -26.440 < 2e-16 ***
SEASON_TICKET_YEARS -0.018959 0.004441 -4.269 1.97e-05 ***
AFFINITY_SCORE 0.159373 0.006827 23.344 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.131517 0.022194 -5.926 3.16e-09 ***
MG_PR_MODEL_DESCMiddle Tier 0.192776 0.028439 6.779 1.25e-11 ***
MG_PR_MODEL_DESCTop Tier 0.567825 0.028286 20.075 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9621 on 18757 degrees of freedom
Multiple R-squared: 0.7257, Adjusted R-squared: 0.7254
F-statistic: 2481 on 20 and 18757 DF, p-value: < 2.2e-16
[[7]][[7]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4118 -0.6261 -0.1835 0.4961 6.2949
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.117126 0.106783 1.097 0.272718
ns(NUMERIC_AGE, df = s)1 0.477331 0.094085 5.073 3.95e-07 ***
ns(NUMERIC_AGE, df = s)2 0.537373 0.119035 4.514 6.39e-06 ***
ns(NUMERIC_AGE, df = s)3 0.392179 0.104574 3.750 0.000177 ***
ns(NUMERIC_AGE, df = s)4 0.246168 0.112685 2.185 0.028932 *
ns(NUMERIC_AGE, df = s)5 -0.266616 0.084922 -3.140 0.001695 **
ns(NUMERIC_AGE, df = s)6 0.767429 0.239493 3.204 0.001356 **
ns(NUMERIC_AGE, df = s)7 -0.489834 0.136162 -3.597 0.000322 ***
PM_VISIT_LAST_2_YRS 0.305753 0.028929 10.569 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.300163 0.028893 10.389 < 2e-16 ***
AF_25K_GIFT 0.624247 0.042920 14.544 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.263555 0.006789 38.821 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.894551 0.042470 68.154 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.565227 0.025606 22.074 < 2e-16 ***
MG_250K_PLUS 1.092212 0.072591 15.046 < 2e-16 ***
Alumnus -0.561957 0.021735 -25.855 < 2e-16 ***
SEASON_TICKET_YEARS -0.018912 0.004532 -4.174 3.01e-05 ***
AFFINITY_SCORE 0.158705 0.006831 23.232 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.108367 0.022213 -4.878 1.08e-06 ***
MG_PR_MODEL_DESCMiddle Tier 0.213574 0.028475 7.501 6.64e-14 ***
MG_PR_MODEL_DESCTop Tier 0.580503 0.028335 20.487 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9634 on 18757 degrees of freedom
Multiple R-squared: 0.725, Adjusted R-squared: 0.7247
F-statistic: 2472 on 20 and 18757 DF, p-value: < 2.2e-16
[[7]][[8]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4619 -0.6286 -0.1796 0.4964 6.2923
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.081983 0.108843 0.753 0.451329
ns(NUMERIC_AGE, df = s)1 0.529849 0.095934 5.523 3.38e-08 ***
ns(NUMERIC_AGE, df = s)2 0.571662 0.121438 4.707 2.53e-06 ***
ns(NUMERIC_AGE, df = s)3 0.459225 0.106672 4.305 1.68e-05 ***
ns(NUMERIC_AGE, df = s)4 0.292024 0.114778 2.544 0.010959 *
ns(NUMERIC_AGE, df = s)5 -0.195408 0.085234 -2.293 0.021882 *
ns(NUMERIC_AGE, df = s)6 0.832848 0.243717 3.417 0.000634 ***
ns(NUMERIC_AGE, df = s)7 -0.516248 0.132786 -3.888 0.000101 ***
PM_VISIT_LAST_2_YRS 0.318383 0.029179 10.911 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.287878 0.029096 9.894 < 2e-16 ***
AF_25K_GIFT 0.578447 0.043677 13.244 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.259340 0.006840 37.916 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.895510 0.042559 68.035 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.553127 0.025767 21.467 < 2e-16 ***
MG_250K_PLUS 1.126523 0.072292 15.583 < 2e-16 ***
Alumnus -0.558337 0.021873 -25.527 < 2e-16 ***
SEASON_TICKET_YEARS -0.018762 0.004484 -4.185 2.87e-05 ***
AFFINITY_SCORE 0.160865 0.006860 23.450 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.116692 0.022257 -5.243 1.60e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.200924 0.028552 7.037 2.03e-12 ***
MG_PR_MODEL_DESCTop Tier 0.586451 0.028420 20.635 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.968 on 18757 degrees of freedom
Multiple R-squared: 0.722, Adjusted R-squared: 0.7217
F-statistic: 2436 on 20 and 18757 DF, p-value: < 2.2e-16
[[7]][[9]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4013 -0.6228 -0.1839 0.4910 6.3317
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.080150 0.107750 0.744 0.45698
ns(NUMERIC_AGE, df = s)1 0.524258 0.094972 5.520 3.43e-08 ***
ns(NUMERIC_AGE, df = s)2 0.555303 0.120203 4.620 3.87e-06 ***
ns(NUMERIC_AGE, df = s)3 0.455051 0.105557 4.311 1.63e-05 ***
ns(NUMERIC_AGE, df = s)4 0.263956 0.113663 2.322 0.02023 *
ns(NUMERIC_AGE, df = s)5 -0.151502 0.085004 -1.782 0.07472 .
ns(NUMERIC_AGE, df = s)6 0.786124 0.241257 3.258 0.00112 **
ns(NUMERIC_AGE, df = s)7 -0.588816 0.134114 -4.390 1.14e-05 ***
PM_VISIT_LAST_2_YRS 0.324652 0.028812 11.268 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.277190 0.028922 9.584 < 2e-16 ***
AF_25K_GIFT 0.570923 0.043056 13.260 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.260401 0.006813 38.223 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.907835 0.042670 68.147 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.573952 0.025623 22.400 < 2e-16 ***
MG_250K_PLUS 1.072566 0.071616 14.977 < 2e-16 ***
Alumnus -0.559668 0.021756 -25.724 < 2e-16 ***
SEASON_TICKET_YEARS -0.022077 0.004460 -4.950 7.47e-07 ***
AFFINITY_SCORE 0.160975 0.006825 23.585 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.118377 0.022294 -5.310 1.11e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.208963 0.028641 7.296 3.09e-13 ***
MG_PR_MODEL_DESCTop Tier 0.590082 0.028307 20.845 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9652 on 18757 degrees of freedom
Multiple R-squared: 0.7257, Adjusted R-squared: 0.7254
F-statistic: 2481 on 20 and 18757 DF, p-value: < 2.2e-16
[[7]][[10]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3582 -0.6307 -0.1864 0.5001 6.3150
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.065436 0.107155 0.611 0.541429
ns(NUMERIC_AGE, df = s)1 0.522871 0.094443 5.536 3.13e-08 ***
ns(NUMERIC_AGE, df = s)2 0.590313 0.119517 4.939 7.91e-07 ***
ns(NUMERIC_AGE, df = s)3 0.461536 0.104972 4.397 1.10e-05 ***
ns(NUMERIC_AGE, df = s)4 0.307790 0.113032 2.723 0.006475 **
ns(NUMERIC_AGE, df = s)5 -0.174227 0.084702 -2.057 0.039707 *
ns(NUMERIC_AGE, df = s)6 0.865603 0.240061 3.606 0.000312 ***
ns(NUMERIC_AGE, df = s)7 -0.564743 0.132928 -4.248 2.16e-05 ***
PM_VISIT_LAST_2_YRS 0.310795 0.028897 10.755 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.284543 0.029051 9.795 < 2e-16 ***
AF_25K_GIFT 0.614032 0.042920 14.306 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.255531 0.006818 37.479 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.913676 0.042640 68.332 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.580105 0.025678 22.592 < 2e-16 ***
MG_250K_PLUS 1.061474 0.071937 14.756 < 2e-16 ***
Alumnus -0.557525 0.021817 -25.554 < 2e-16 ***
SEASON_TICKET_YEARS -0.021370 0.004513 -4.736 2.20e-06 ***
AFFINITY_SCORE 0.162192 0.006838 23.718 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.111517 0.022350 -4.990 6.11e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.204501 0.028618 7.146 9.27e-13 ***
MG_PR_MODEL_DESCTop Tier 0.579304 0.028415 20.387 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9672 on 18753 degrees of freedom
Multiple R-squared: 0.7233, Adjusted R-squared: 0.7231
F-statistic: 2452 on 20 and 18753 DF, p-value: < 2.2e-16
[[8]]
[[8]][[1]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4360 -0.6239 -0.1774 0.4913 5.4687
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.039417 0.113821 0.346 0.72911
ns(NUMERIC_AGE, df = s)1 0.585910 0.102169 5.735 9.92e-09 ***
ns(NUMERIC_AGE, df = s)2 0.687613 0.129206 5.322 1.04e-07 ***
ns(NUMERIC_AGE, df = s)3 0.510361 0.114241 4.467 7.96e-06 ***
ns(NUMERIC_AGE, df = s)4 0.551116 0.121474 4.537 5.74e-06 ***
ns(NUMERIC_AGE, df = s)5 0.265351 0.119434 2.222 0.02631 *
ns(NUMERIC_AGE, df = s)6 -0.082664 0.090404 -0.914 0.36053
ns(NUMERIC_AGE, df = s)7 0.839434 0.255648 3.284 0.00103 **
ns(NUMERIC_AGE, df = s)8 -0.718695 0.142863 -5.031 4.93e-07 ***
PM_VISIT_LAST_2_YRS 0.312134 0.028727 10.865 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.274596 0.028870 9.511 < 2e-16 ***
AF_25K_GIFT 0.616388 0.042553 14.485 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.261012 0.006781 38.493 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.928066 0.042353 69.135 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.557704 0.025517 21.856 < 2e-16 ***
MG_250K_PLUS 1.127244 0.071594 15.745 < 2e-16 ***
Alumnus -0.568534 0.021779 -26.104 < 2e-16 ***
SEASON_TICKET_YEARS -0.019413 0.004460 -4.353 1.35e-05 ***
AFFINITY_SCORE 0.158313 0.006809 23.250 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.117186 0.022122 -5.297 1.19e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.215217 0.028342 7.593 3.26e-14 ***
MG_PR_MODEL_DESCTop Tier 0.596246 0.028194 21.148 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9607 on 18756 degrees of freedom
Multiple R-squared: 0.7273, Adjusted R-squared: 0.727
F-statistic: 2382 on 21 and 18756 DF, p-value: < 2.2e-16
[[8]][[2]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4810 -0.6225 -0.1721 0.4904 6.3360
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.040003 0.112265 0.356 0.721597
ns(NUMERIC_AGE, df = s)1 0.571004 0.100752 5.667 1.47e-08 ***
ns(NUMERIC_AGE, df = s)2 0.683037 0.127519 5.356 8.59e-08 ***
ns(NUMERIC_AGE, df = s)3 0.531198 0.112442 4.724 2.33e-06 ***
ns(NUMERIC_AGE, df = s)4 0.525181 0.120921 4.343 1.41e-05 ***
ns(NUMERIC_AGE, df = s)5 0.260089 0.118274 2.199 0.027887 *
ns(NUMERIC_AGE, df = s)6 -0.103058 0.087377 -1.179 0.238228
ns(NUMERIC_AGE, df = s)7 0.911108 0.251292 3.626 0.000289 ***
ns(NUMERIC_AGE, df = s)8 -0.624626 0.130205 -4.797 1.62e-06 ***
PM_VISIT_LAST_2_YRS 0.302816 0.028782 10.521 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.279913 0.028950 9.669 < 2e-16 ***
AF_25K_GIFT 0.588782 0.043183 13.635 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.259966 0.006767 38.415 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.911550 0.042363 68.729 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.556767 0.025554 21.788 < 2e-16 ***
MG_250K_PLUS 1.167345 0.070022 16.671 < 2e-16 ***
Alumnus -0.563052 0.021840 -25.781 < 2e-16 ***
SEASON_TICKET_YEARS -0.018879 0.004504 -4.192 2.78e-05 ***
AFFINITY_SCORE 0.158868 0.006805 23.347 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.122321 0.022100 -5.535 3.15e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.208697 0.028340 7.364 1.86e-13 ***
MG_PR_MODEL_DESCTop Tier 0.601444 0.028339 21.223 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9606 on 18756 degrees of freedom
Multiple R-squared: 0.7276, Adjusted R-squared: 0.7273
F-statistic: 2386 on 21 and 18756 DF, p-value: < 2.2e-16
[[8]][[3]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4051 -0.6246 -0.1799 0.4914 6.3571
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.101223 0.115722 0.875 0.381744
ns(NUMERIC_AGE, df = s)1 0.511486 0.103752 4.930 8.30e-07 ***
ns(NUMERIC_AGE, df = s)2 0.621729 0.131482 4.729 2.28e-06 ***
ns(NUMERIC_AGE, df = s)3 0.451576 0.116240 3.885 0.000103 ***
ns(NUMERIC_AGE, df = s)4 0.462241 0.123461 3.744 0.000182 ***
ns(NUMERIC_AGE, df = s)5 0.192943 0.121351 1.590 0.111862
ns(NUMERIC_AGE, df = s)6 -0.160660 0.090896 -1.768 0.077157 .
ns(NUMERIC_AGE, df = s)7 0.754015 0.259879 2.901 0.003719 **
ns(NUMERIC_AGE, df = s)8 -0.649946 0.142084 -4.574 4.81e-06 ***
PM_VISIT_LAST_2_YRS 0.320887 0.028788 11.147 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.263522 0.028884 9.124 < 2e-16 ***
AF_25K_GIFT 0.614004 0.042748 14.363 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.263089 0.006793 38.728 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.913994 0.042465 68.621 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.562786 0.025655 21.936 < 2e-16 ***
MG_250K_PLUS 1.090803 0.071099 15.342 < 2e-16 ***
Alumnus -0.570637 0.021839 -26.130 < 2e-16 ***
SEASON_TICKET_YEARS -0.020240 0.004419 -4.580 4.67e-06 ***
AFFINITY_SCORE 0.160614 0.006835 23.498 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.109727 0.022171 -4.949 7.52e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.207893 0.028455 7.306 2.86e-13 ***
MG_PR_MODEL_DESCTop Tier 0.596034 0.028346 21.027 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9637 on 18756 degrees of freedom
Multiple R-squared: 0.7262, Adjusted R-squared: 0.7259
F-statistic: 2369 on 21 and 18756 DF, p-value: < 2.2e-16
[[8]][[4]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4087 -0.6246 -0.1777 0.4929 6.3194
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.036173 0.113345 0.319 0.749625
ns(NUMERIC_AGE, df = s)1 0.593737 0.101664 5.840 5.30e-09 ***
ns(NUMERIC_AGE, df = s)2 0.698901 0.128636 5.433 5.61e-08 ***
ns(NUMERIC_AGE, df = s)3 0.541457 0.113456 4.772 1.83e-06 ***
ns(NUMERIC_AGE, df = s)4 0.545744 0.121760 4.482 7.43e-06 ***
ns(NUMERIC_AGE, df = s)5 0.275220 0.119179 2.309 0.020938 *
ns(NUMERIC_AGE, df = s)6 -0.125082 0.089682 -1.395 0.163113
ns(NUMERIC_AGE, df = s)7 0.970271 0.254893 3.807 0.000141 ***
ns(NUMERIC_AGE, df = s)8 -0.575917 0.145263 -3.965 7.38e-05 ***
PM_VISIT_LAST_2_YRS 0.320326 0.028951 11.064 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.288626 0.028917 9.981 < 2e-16 ***
AF_25K_GIFT 0.626123 0.042881 14.601 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.259639 0.006801 38.177 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.923620 0.042426 68.912 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.577779 0.025548 22.615 < 2e-16 ***
MG_250K_PLUS 1.092778 0.071361 15.313 < 2e-16 ***
Alumnus -0.568595 0.021860 -26.011 < 2e-16 ***
SEASON_TICKET_YEARS -0.019796 0.004441 -4.457 8.35e-06 ***
AFFINITY_SCORE 0.155950 0.006841 22.796 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.123396 0.022218 -5.554 2.83e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.193508 0.028528 6.783 1.21e-11 ***
MG_PR_MODEL_DESCTop Tier 0.577802 0.028381 20.359 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9644 on 18756 degrees of freedom
Multiple R-squared: 0.7247, Adjusted R-squared: 0.7244
F-statistic: 2351 on 21 and 18756 DF, p-value: < 2.2e-16
[[8]][[5]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.1028 -0.6258 -0.1788 0.4967 6.3397
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.078849 0.114797 0.687 0.49218
ns(NUMERIC_AGE, df = s)1 0.549310 0.102926 5.337 9.56e-08 ***
ns(NUMERIC_AGE, df = s)2 0.649338 0.130328 4.982 6.34e-07 ***
ns(NUMERIC_AGE, df = s)3 0.486487 0.114948 4.232 2.32e-05 ***
ns(NUMERIC_AGE, df = s)4 0.509428 0.123351 4.130 3.65e-05 ***
ns(NUMERIC_AGE, df = s)5 0.203694 0.120602 1.689 0.09124 .
ns(NUMERIC_AGE, df = s)6 -0.090804 0.090411 -1.004 0.31522
ns(NUMERIC_AGE, df = s)7 0.809044 0.258494 3.130 0.00175 **
ns(NUMERIC_AGE, df = s)8 -0.679541 0.147024 -4.622 3.83e-06 ***
PM_VISIT_LAST_2_YRS 0.339835 0.029080 11.686 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.278136 0.029125 9.550 < 2e-16 ***
AF_25K_GIFT 0.607513 0.043039 14.116 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.258800 0.006806 38.024 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.915493 0.042638 68.378 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.558145 0.025740 21.684 < 2e-16 ***
MG_250K_PLUS 1.098893 0.071850 15.294 < 2e-16 ***
Alumnus -0.568868 0.021958 -25.907 < 2e-16 ***
SEASON_TICKET_YEARS -0.020895 0.004514 -4.629 3.70e-06 ***
AFFINITY_SCORE 0.158911 0.006857 23.176 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.115917 0.022296 -5.199 2.02e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.220942 0.028519 7.747 9.87e-15 ***
MG_PR_MODEL_DESCTop Tier 0.595249 0.028410 20.952 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9677 on 18756 degrees of freedom
Multiple R-squared: 0.7233, Adjusted R-squared: 0.723
F-statistic: 2335 on 21 and 18756 DF, p-value: < 2.2e-16
[[8]][[6]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4449 -0.6200 -0.1781 0.4932 6.3541
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.033703 0.112716 0.299 0.76493
ns(NUMERIC_AGE, df = s)1 0.614633 0.101090 6.080 1.22e-09 ***
ns(NUMERIC_AGE, df = s)2 0.746538 0.128159 5.825 5.80e-09 ***
ns(NUMERIC_AGE, df = s)3 0.554053 0.112870 4.909 9.24e-07 ***
ns(NUMERIC_AGE, df = s)4 0.550651 0.121208 4.543 5.58e-06 ***
ns(NUMERIC_AGE, df = s)5 0.274541 0.118571 2.315 0.02060 *
ns(NUMERIC_AGE, df = s)6 -0.088302 0.089561 -0.986 0.32417
ns(NUMERIC_AGE, df = s)7 0.935473 0.253931 3.684 0.00023 ***
ns(NUMERIC_AGE, df = s)8 -0.670750 0.146112 -4.591 4.45e-06 ***
PM_VISIT_LAST_2_YRS 0.285080 0.029121 9.790 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.301734 0.028875 10.450 < 2e-16 ***
AF_25K_GIFT 0.614958 0.043462 14.149 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.260416 0.006792 38.342 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.912006 0.042483 68.545 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.561857 0.025558 21.984 < 2e-16 ***
MG_250K_PLUS 1.141667 0.072210 15.810 < 2e-16 ***
Alumnus -0.577741 0.021762 -26.548 < 2e-16 ***
SEASON_TICKET_YEARS -0.019178 0.004441 -4.319 1.58e-05 ***
AFFINITY_SCORE 0.158678 0.006831 23.228 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.131489 0.022191 -5.925 3.17e-09 ***
MG_PR_MODEL_DESCMiddle Tier 0.192268 0.028434 6.762 1.40e-11 ***
MG_PR_MODEL_DESCTop Tier 0.567385 0.028282 20.062 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.962 on 18756 degrees of freedom
Multiple R-squared: 0.7258, Adjusted R-squared: 0.7255
F-statistic: 2364 on 21 and 18756 DF, p-value: < 2.2e-16
[[8]][[7]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4059 -0.6252 -0.1812 0.4954 6.3254
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.036317 0.112275 0.323 0.746349
ns(NUMERIC_AGE, df = s)1 0.560646 0.100559 5.575 2.51e-08 ***
ns(NUMERIC_AGE, df = s)2 0.717072 0.127574 5.621 1.93e-08 ***
ns(NUMERIC_AGE, df = s)3 0.500916 0.112321 4.460 8.26e-06 ***
ns(NUMERIC_AGE, df = s)4 0.553400 0.120852 4.579 4.70e-06 ***
ns(NUMERIC_AGE, df = s)5 0.234639 0.118137 1.986 0.047030 *
ns(NUMERIC_AGE, df = s)6 -0.141585 0.089535 -1.581 0.113816
ns(NUMERIC_AGE, df = s)7 0.937660 0.253042 3.706 0.000212 ***
ns(NUMERIC_AGE, df = s)8 -0.596599 0.148405 -4.020 5.84e-05 ***
PM_VISIT_LAST_2_YRS 0.307266 0.028926 10.623 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.299547 0.028885 10.370 < 2e-16 ***
AF_25K_GIFT 0.623457 0.042910 14.530 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.263117 0.006788 38.759 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.898624 0.042475 68.243 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.567138 0.025604 22.150 < 2e-16 ***
MG_250K_PLUS 1.095351 0.072576 15.093 < 2e-16 ***
Alumnus -0.569093 0.021847 -26.049 < 2e-16 ***
SEASON_TICKET_YEARS -0.019124 0.004531 -4.221 2.44e-05 ***
AFFINITY_SCORE 0.157861 0.006834 23.099 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.108504 0.022208 -4.886 1.04e-06 ***
MG_PR_MODEL_DESCMiddle Tier 0.213360 0.028467 7.495 6.92e-14 ***
MG_PR_MODEL_DESCTop Tier 0.580478 0.028328 20.491 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9632 on 18756 degrees of freedom
Multiple R-squared: 0.7251, Adjusted R-squared: 0.7248
F-statistic: 2356 on 21 and 18756 DF, p-value: < 2.2e-16
[[8]][[8]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4564 -0.6278 -0.1766 0.4946 6.3215
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.021564 0.114573 0.188 0.850711
ns(NUMERIC_AGE, df = s)1 0.607043 0.102690 5.911 3.45e-09 ***
ns(NUMERIC_AGE, df = s)2 0.707415 0.130058 5.439 5.42e-08 ***
ns(NUMERIC_AGE, df = s)3 0.537645 0.114753 4.685 2.82e-06 ***
ns(NUMERIC_AGE, df = s)4 0.590018 0.123170 4.790 1.68e-06 ***
ns(NUMERIC_AGE, df = s)5 0.260681 0.120440 2.164 0.030446 *
ns(NUMERIC_AGE, df = s)6 -0.076845 0.090046 -0.853 0.393446
ns(NUMERIC_AGE, df = s)7 0.940034 0.257719 3.648 0.000266 ***
ns(NUMERIC_AGE, df = s)8 -0.623593 0.144438 -4.317 1.59e-05 ***
PM_VISIT_LAST_2_YRS 0.319424 0.029177 10.948 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.287294 0.029090 9.876 < 2e-16 ***
AF_25K_GIFT 0.577241 0.043671 13.218 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.258946 0.006840 37.857 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.899435 0.042572 68.107 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.555049 0.025772 21.537 < 2e-16 ***
MG_250K_PLUS 1.129104 0.072280 15.621 < 2e-16 ***
Alumnus -0.563961 0.021985 -25.652 < 2e-16 ***
SEASON_TICKET_YEARS -0.018942 0.004483 -4.225 2.40e-05 ***
AFFINITY_SCORE 0.160015 0.006865 23.307 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.116747 0.022253 -5.246 1.57e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.200708 0.028546 7.031 2.12e-12 ***
MG_PR_MODEL_DESCTop Tier 0.586574 0.028416 20.643 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9678 on 18756 degrees of freedom
Multiple R-squared: 0.7221, Adjusted R-squared: 0.7218
F-statistic: 2321 on 21 and 18756 DF, p-value: < 2.2e-16
[[8]][[9]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3971 -0.6232 -0.1815 0.4909 6.3536
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.027123 0.113380 0.239 0.810939
ns(NUMERIC_AGE, df = s)1 0.597621 0.101547 5.885 4.04e-09 ***
ns(NUMERIC_AGE, df = s)2 0.673127 0.129078 5.215 1.86e-07 ***
ns(NUMERIC_AGE, df = s)3 0.528538 0.113797 4.645 3.43e-06 ***
ns(NUMERIC_AGE, df = s)4 0.557168 0.121192 4.597 4.31e-06 ***
ns(NUMERIC_AGE, df = s)5 0.252020 0.119091 2.116 0.034342 *
ns(NUMERIC_AGE, df = s)6 -0.056571 0.090606 -0.624 0.532399
ns(NUMERIC_AGE, df = s)7 0.874135 0.254922 3.429 0.000607 ***
ns(NUMERIC_AGE, df = s)8 -0.667931 0.144264 -4.630 3.68e-06 ***
PM_VISIT_LAST_2_YRS 0.325656 0.028812 11.303 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.276726 0.028919 9.569 < 2e-16 ***
AF_25K_GIFT 0.570566 0.043052 13.253 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.260012 0.006814 38.158 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.911048 0.042689 68.192 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.575136 0.025625 22.444 < 2e-16 ***
MG_250K_PLUS 1.074767 0.071610 15.009 < 2e-16 ***
Alumnus -0.564220 0.021862 -25.808 < 2e-16 ***
SEASON_TICKET_YEARS -0.022254 0.004460 -4.990 6.10e-07 ***
AFFINITY_SCORE 0.160432 0.006829 23.494 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.118532 0.022292 -5.317 1.07e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.208829 0.028638 7.292 3.17e-13 ***
MG_PR_MODEL_DESCTop Tier 0.589972 0.028306 20.843 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.965 on 18756 degrees of freedom
Multiple R-squared: 0.7257, Adjusted R-squared: 0.7254
F-statistic: 2363 on 21 and 18756 DF, p-value: < 2.2e-16
[[8]][[10]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3523 -0.6296 -0.1818 0.4960 6.3432
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.008138 0.112603 -0.072 0.9424
ns(NUMERIC_AGE, df = s)1 0.609698 0.100947 6.040 1.57e-09 ***
ns(NUMERIC_AGE, df = s)2 0.743336 0.127997 5.807 6.45e-09 ***
ns(NUMERIC_AGE, df = s)3 0.561059 0.112700 4.978 6.47e-07 ***
ns(NUMERIC_AGE, df = s)4 0.603178 0.121048 4.983 6.32e-07 ***
ns(NUMERIC_AGE, df = s)5 0.297353 0.118473 2.510 0.0121 *
ns(NUMERIC_AGE, df = s)6 -0.061400 0.089399 -0.687 0.4922
ns(NUMERIC_AGE, df = s)7 1.007888 0.253450 3.977 7.01e-05 ***
ns(NUMERIC_AGE, df = s)8 -0.663869 0.144642 -4.590 4.47e-06 ***
PM_VISIT_LAST_2_YRS 0.312375 0.028896 10.810 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.284032 0.029045 9.779 < 2e-16 ***
AF_25K_GIFT 0.613419 0.042912 14.295 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.255119 0.006818 37.419 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.917870 0.042654 68.407 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.582135 0.025683 22.666 < 2e-16 ***
MG_250K_PLUS 1.063860 0.071922 14.792 < 2e-16 ***
Alumnus -0.563594 0.021929 -25.700 < 2e-16 ***
SEASON_TICKET_YEARS -0.021563 0.004512 -4.779 1.78e-06 ***
AFFINITY_SCORE 0.161381 0.006842 23.585 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.111513 0.022345 -4.990 6.08e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.204314 0.028612 7.141 9.61e-13 ***
MG_PR_MODEL_DESCTop Tier 0.579105 0.028410 20.384 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.967 on 18752 degrees of freedom
Multiple R-squared: 0.7235, Adjusted R-squared: 0.7232
F-statistic: 2336 on 21 and 18752 DF, p-value: < 2.2e-16
[[9]]
[[9]][[1]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4335 -0.6243 -0.1774 0.4905 5.4701
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.043233 0.116657 0.371 0.71094
ns(NUMERIC_AGE, df = s)1 0.609864 0.104591 5.831 5.60e-09 ***
ns(NUMERIC_AGE, df = s)2 0.659995 0.136859 4.822 1.43e-06 ***
ns(NUMERIC_AGE, df = s)3 0.569488 0.122633 4.644 3.44e-06 ***
ns(NUMERIC_AGE, df = s)4 0.508538 0.117941 4.312 1.63e-05 ***
ns(NUMERIC_AGE, df = s)5 0.533451 0.123794 4.309 1.65e-05 ***
ns(NUMERIC_AGE, df = s)6 0.232999 0.123264 1.890 0.05874 .
ns(NUMERIC_AGE, df = s)7 -0.064007 0.091556 -0.699 0.48450
ns(NUMERIC_AGE, df = s)8 0.804702 0.262519 3.065 0.00218 **
ns(NUMERIC_AGE, df = s)9 -0.757331 0.150135 -5.044 4.59e-07 ***
PM_VISIT_LAST_2_YRS 0.312065 0.028729 10.862 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.274254 0.028872 9.499 < 2e-16 ***
AF_25K_GIFT 0.616623 0.042554 14.490 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.260929 0.006782 38.472 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.929151 0.042386 69.107 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.558113 0.025528 21.863 < 2e-16 ***
MG_250K_PLUS 1.127590 0.071596 15.749 < 2e-16 ***
Alumnus -0.570097 0.021959 -25.962 < 2e-16 ***
SEASON_TICKET_YEARS -0.019403 0.004460 -4.350 1.37e-05 ***
AFFINITY_SCORE 0.158088 0.006818 23.186 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.117472 0.022124 -5.310 1.11e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.215304 0.028344 7.596 3.19e-14 ***
MG_PR_MODEL_DESCTop Tier 0.596040 0.028194 21.140 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9607 on 18755 degrees of freedom
Multiple R-squared: 0.7273, Adjusted R-squared: 0.727
F-statistic: 2274 on 22 and 18755 DF, p-value: < 2.2e-16
[[9]][[2]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4783 -0.6220 -0.1736 0.4908 6.3405
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.046815 0.118156 0.396 0.69195
ns(NUMERIC_AGE, df = s)1 0.623539 0.106234 5.869 4.45e-09 ***
ns(NUMERIC_AGE, df = s)2 0.614043 0.136613 4.495 7.01e-06 ***
ns(NUMERIC_AGE, df = s)3 0.624711 0.124631 5.013 5.42e-07 ***
ns(NUMERIC_AGE, df = s)4 0.505469 0.118920 4.250 2.14e-05 ***
ns(NUMERIC_AGE, df = s)5 0.507807 0.125355 4.051 5.12e-05 ***
ns(NUMERIC_AGE, df = s)6 0.237740 0.124735 1.906 0.05667 .
ns(NUMERIC_AGE, df = s)7 -0.081224 0.090229 -0.900 0.36803
ns(NUMERIC_AGE, df = s)8 0.857762 0.264549 3.242 0.00119 **
ns(NUMERIC_AGE, df = s)9 -0.629725 0.135338 -4.653 3.29e-06 ***
PM_VISIT_LAST_2_YRS 0.302704 0.028783 10.517 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.279857 0.028952 9.666 < 2e-16 ***
AF_25K_GIFT 0.588696 0.043186 13.632 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.259856 0.006769 38.387 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.912615 0.042399 68.695 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.557303 0.025569 21.796 < 2e-16 ***
MG_250K_PLUS 1.167038 0.070031 16.665 < 2e-16 ***
Alumnus -0.564550 0.022016 -25.643 < 2e-16 ***
SEASON_TICKET_YEARS -0.018829 0.004504 -4.180 2.92e-05 ***
AFFINITY_SCORE 0.158626 0.006815 23.275 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.122681 0.022103 -5.550 2.89e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.208785 0.028342 7.367 1.82e-13 ***
MG_PR_MODEL_DESCTop Tier 0.601293 0.028341 21.217 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9606 on 18755 degrees of freedom
Multiple R-squared: 0.7276, Adjusted R-squared: 0.7273
F-statistic: 2277 on 22 and 18755 DF, p-value: < 2.2e-16
[[9]][[3]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4010 -0.6236 -0.1796 0.4930 6.3656
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.109297 0.122011 0.896 0.370374
ns(NUMERIC_AGE, df = s)1 0.559844 0.109655 5.106 3.33e-07 ***
ns(NUMERIC_AGE, df = s)2 0.555481 0.140669 3.949 7.88e-05 ***
ns(NUMERIC_AGE, df = s)3 0.554946 0.128305 4.325 1.53e-05 ***
ns(NUMERIC_AGE, df = s)4 0.428033 0.122885 3.483 0.000497 ***
ns(NUMERIC_AGE, df = s)5 0.447244 0.128947 3.468 0.000525 ***
ns(NUMERIC_AGE, df = s)6 0.155074 0.128345 1.208 0.226963
ns(NUMERIC_AGE, df = s)7 -0.132127 0.093101 -1.419 0.155863
ns(NUMERIC_AGE, df = s)8 0.698768 0.274223 2.548 0.010837 *
ns(NUMERIC_AGE, df = s)9 -0.676936 0.149575 -4.526 6.06e-06 ***
PM_VISIT_LAST_2_YRS 0.320899 0.028789 11.147 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.263377 0.028884 9.118 < 2e-16 ***
AF_25K_GIFT 0.613935 0.042749 14.361 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.262919 0.006795 38.691 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.915772 0.042501 68.605 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.563544 0.025669 21.954 < 2e-16 ***
MG_250K_PLUS 1.090876 0.071105 15.342 < 2e-16 ***
Alumnus -0.573009 0.022006 -26.039 < 2e-16 ***
SEASON_TICKET_YEARS -0.020194 0.004419 -4.570 4.91e-06 ***
AFFINITY_SCORE 0.160243 0.006845 23.409 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.110136 0.022172 -4.967 6.85e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.207916 0.028456 7.307 2.85e-13 ***
MG_PR_MODEL_DESCTop Tier 0.595628 0.028348 21.011 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9637 on 18755 degrees of freedom
Multiple R-squared: 0.7262, Adjusted R-squared: 0.7259
F-statistic: 2261 on 22 and 18755 DF, p-value: < 2.2e-16
[[9]][[4]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4047 -0.6254 -0.1797 0.4936 6.3255
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.038428 0.118634 0.324 0.746004
ns(NUMERIC_AGE, df = s)1 0.639086 0.106217 6.017 1.81e-09 ***
ns(NUMERIC_AGE, df = s)2 0.648075 0.137655 4.708 2.52e-06 ***
ns(NUMERIC_AGE, df = s)3 0.630982 0.124249 5.078 3.84e-07 ***
ns(NUMERIC_AGE, df = s)4 0.519313 0.119884 4.332 1.49e-05 ***
ns(NUMERIC_AGE, df = s)5 0.546987 0.125670 4.353 1.35e-05 ***
ns(NUMERIC_AGE, df = s)6 0.250869 0.125188 2.004 0.045090 *
ns(NUMERIC_AGE, df = s)7 -0.091087 0.092550 -0.984 0.325036
ns(NUMERIC_AGE, df = s)8 0.930163 0.266476 3.491 0.000483 ***
ns(NUMERIC_AGE, df = s)9 -0.592980 0.151347 -3.918 8.96e-05 ***
PM_VISIT_LAST_2_YRS 0.320340 0.028951 11.065 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.288493 0.028918 9.976 < 2e-16 ***
AF_25K_GIFT 0.625901 0.042883 14.596 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.259445 0.006803 38.136 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.925387 0.042457 68.903 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.578563 0.025560 22.636 < 2e-16 ***
MG_250K_PLUS 1.093101 0.071362 15.318 < 2e-16 ***
Alumnus -0.571320 0.022021 -25.944 < 2e-16 ***
SEASON_TICKET_YEARS -0.019771 0.004441 -4.452 8.57e-06 ***
AFFINITY_SCORE 0.155581 0.006850 22.713 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.123681 0.022219 -5.566 2.63e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.193725 0.028529 6.791 1.15e-11 ***
MG_PR_MODEL_DESCTop Tier 0.577613 0.028382 20.351 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9644 on 18755 degrees of freedom
Multiple R-squared: 0.7247, Adjusted R-squared: 0.7244
F-statistic: 2244 on 22 and 18755 DF, p-value: < 2.2e-16
[[9]][[5]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.0945 -0.6265 -0.1794 0.4966 6.3468
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.080066 0.120305 0.666 0.50572
ns(NUMERIC_AGE, df = s)1 0.589559 0.107668 5.476 4.41e-08 ***
ns(NUMERIC_AGE, df = s)2 0.606487 0.139588 4.345 1.40e-05 ***
ns(NUMERIC_AGE, df = s)3 0.566900 0.125829 4.505 6.67e-06 ***
ns(NUMERIC_AGE, df = s)4 0.474504 0.121674 3.900 9.66e-05 ***
ns(NUMERIC_AGE, df = s)5 0.504526 0.127373 3.961 7.49e-05 ***
ns(NUMERIC_AGE, df = s)6 0.182344 0.126832 1.438 0.15054
ns(NUMERIC_AGE, df = s)7 -0.058585 0.093305 -0.628 0.53009
ns(NUMERIC_AGE, df = s)8 0.769905 0.270527 2.846 0.00443 **
ns(NUMERIC_AGE, df = s)9 -0.702851 0.153113 -4.590 4.45e-06 ***
PM_VISIT_LAST_2_YRS 0.339893 0.029081 11.688 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.277844 0.029127 9.539 < 2e-16 ***
AF_25K_GIFT 0.607484 0.043040 14.114 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.258648 0.006809 37.988 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.917173 0.042674 68.359 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.558778 0.025752 21.699 < 2e-16 ***
MG_250K_PLUS 1.099080 0.071854 15.296 < 2e-16 ***
Alumnus -0.571095 0.022123 -25.814 < 2e-16 ***
SEASON_TICKET_YEARS -0.020843 0.004514 -4.618 3.91e-06 ***
AFFINITY_SCORE 0.158578 0.006867 23.092 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.116192 0.022297 -5.211 1.90e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.221126 0.028521 7.753 9.42e-15 ***
MG_PR_MODEL_DESCTop Tier 0.595143 0.028411 20.948 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9678 on 18755 degrees of freedom
Multiple R-squared: 0.7233, Adjusted R-squared: 0.723
F-statistic: 2228 on 22 and 18755 DF, p-value: < 2.2e-16
[[9]][[6]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4434 -0.6203 -0.1780 0.4929 6.3594
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.022080 0.118074 0.187 0.851661
ns(NUMERIC_AGE, df = s)1 0.659209 0.105666 6.239 4.51e-10 ***
ns(NUMERIC_AGE, df = s)2 0.726015 0.137284 5.288 1.25e-07 ***
ns(NUMERIC_AGE, df = s)3 0.629364 0.123736 5.086 3.69e-07 ***
ns(NUMERIC_AGE, df = s)4 0.550476 0.119410 4.610 4.05e-06 ***
ns(NUMERIC_AGE, df = s)5 0.549707 0.125132 4.393 1.12e-05 ***
ns(NUMERIC_AGE, df = s)6 0.268880 0.124656 2.157 0.031020 *
ns(NUMERIC_AGE, df = s)7 -0.058236 0.092449 -0.630 0.528751
ns(NUMERIC_AGE, df = s)8 0.925741 0.265669 3.485 0.000494 ***
ns(NUMERIC_AGE, df = s)9 -0.683288 0.152317 -4.486 7.30e-06 ***
PM_VISIT_LAST_2_YRS 0.284848 0.029122 9.781 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.301671 0.028877 10.447 < 2e-16 ***
AF_25K_GIFT 0.614871 0.043465 14.147 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.260323 0.006794 38.315 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.912918 0.042524 68.501 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.562144 0.025570 21.985 < 2e-16 ***
MG_250K_PLUS 1.142265 0.072215 15.818 < 2e-16 ***
Alumnus -0.578818 0.021937 -26.386 < 2e-16 ***
SEASON_TICKET_YEARS -0.019138 0.004441 -4.310 1.64e-05 ***
AFFINITY_SCORE 0.158529 0.006842 23.171 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.131673 0.022192 -5.933 3.02e-09 ***
MG_PR_MODEL_DESCMiddle Tier 0.192290 0.028436 6.762 1.40e-11 ***
MG_PR_MODEL_DESCTop Tier 0.567190 0.028283 20.054 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.962 on 18755 degrees of freedom
Multiple R-squared: 0.7258, Adjusted R-squared: 0.7255
F-statistic: 2256 on 22 and 18755 DF, p-value: < 2.2e-16
[[9]][[7]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4047 -0.6251 -0.1818 0.4961 6.3309
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.021401 0.118198 0.181 0.856320
ns(NUMERIC_AGE, df = s)1 0.621346 0.106059 5.858 4.75e-09 ***
ns(NUMERIC_AGE, df = s)2 0.685163 0.136634 5.015 5.36e-07 ***
ns(NUMERIC_AGE, df = s)3 0.606448 0.124526 4.870 1.12e-06 ***
ns(NUMERIC_AGE, df = s)4 0.518609 0.118872 4.363 1.29e-05 ***
ns(NUMERIC_AGE, df = s)5 0.554099 0.125323 4.421 9.86e-06 ***
ns(NUMERIC_AGE, df = s)6 0.228227 0.124679 1.831 0.067188 .
ns(NUMERIC_AGE, df = s)7 -0.106242 0.092407 -1.150 0.250272
ns(NUMERIC_AGE, df = s)8 0.938025 0.266728 3.517 0.000438 ***
ns(NUMERIC_AGE, df = s)9 -0.610726 0.154684 -3.948 7.90e-05 ***
PM_VISIT_LAST_2_YRS 0.306956 0.028928 10.611 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.299472 0.028887 10.367 < 2e-16 ***
AF_25K_GIFT 0.623342 0.042913 14.526 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.263047 0.006791 38.736 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.899187 0.042517 68.188 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.567433 0.025620 22.148 < 2e-16 ***
MG_250K_PLUS 1.095916 0.072585 15.098 < 2e-16 ***
Alumnus -0.569906 0.022009 -25.894 < 2e-16 ***
SEASON_TICKET_YEARS -0.019078 0.004531 -4.211 2.56e-05 ***
AFFINITY_SCORE 0.157738 0.006846 23.042 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.108838 0.022210 -4.900 9.65e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.213298 0.028469 7.492 7.07e-14 ***
MG_PR_MODEL_DESCTop Tier 0.580252 0.028330 20.482 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9632 on 18755 degrees of freedom
Multiple R-squared: 0.7251, Adjusted R-squared: 0.7248
F-statistic: 2249 on 22 and 18755 DF, p-value: < 2.2e-16
[[9]][[8]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4515 -0.6272 -0.1762 0.4941 6.3301
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.031027 0.120668 0.257 0.79708
ns(NUMERIC_AGE, df = s)1 0.657012 0.108424 6.060 1.39e-09 ***
ns(NUMERIC_AGE, df = s)2 0.639457 0.139258 4.592 4.42e-06 ***
ns(NUMERIC_AGE, df = s)3 0.645665 0.127182 5.077 3.88e-07 ***
ns(NUMERIC_AGE, df = s)4 0.524711 0.121506 4.318 1.58e-05 ***
ns(NUMERIC_AGE, df = s)5 0.585682 0.127770 4.584 4.59e-06 ***
ns(NUMERIC_AGE, df = s)6 0.224496 0.127126 1.766 0.07742 .
ns(NUMERIC_AGE, df = s)7 -0.037044 0.092985 -0.398 0.69035
ns(NUMERIC_AGE, df = s)8 0.877655 0.271614 3.231 0.00123 **
ns(NUMERIC_AGE, df = s)9 -0.654601 0.150392 -4.353 1.35e-05 ***
PM_VISIT_LAST_2_YRS 0.319466 0.029177 10.949 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.287100 0.029091 9.869 < 2e-16 ***
AF_25K_GIFT 0.576832 0.043673 13.208 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.258713 0.006843 37.810 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.901738 0.042615 68.092 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.555890 0.025784 21.560 < 2e-16 ***
MG_250K_PLUS 1.129077 0.072282 15.620 < 2e-16 ***
Alumnus -0.566801 0.022143 -25.597 < 2e-16 ***
SEASON_TICKET_YEARS -0.018891 0.004483 -4.214 2.52e-05 ***
AFFINITY_SCORE 0.159587 0.006875 23.212 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.117186 0.022253 -5.266 1.41e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.200778 0.028547 7.033 2.09e-12 ***
MG_PR_MODEL_DESCTop Tier 0.586304 0.028415 20.633 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9678 on 18755 degrees of freedom
Multiple R-squared: 0.7222, Adjusted R-squared: 0.7218
F-statistic: 2216 on 22 and 18755 DF, p-value: < 2.2e-16
[[9]][[9]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3947 -0.6230 -0.1812 0.4884 6.3617
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.029109 0.119431 0.244 0.80744
ns(NUMERIC_AGE, df = s)1 0.645416 0.107136 6.024 1.73e-09 ***
ns(NUMERIC_AGE, df = s)2 0.630709 0.138170 4.565 5.03e-06 ***
ns(NUMERIC_AGE, df = s)3 0.608862 0.125883 4.837 1.33e-06 ***
ns(NUMERIC_AGE, df = s)4 0.523269 0.120154 4.355 1.34e-05 ***
ns(NUMERIC_AGE, df = s)5 0.533972 0.126523 4.220 2.45e-05 ***
ns(NUMERIC_AGE, df = s)6 0.226269 0.125820 1.798 0.07214 .
ns(NUMERIC_AGE, df = s)7 -0.031000 0.092771 -0.334 0.73826
ns(NUMERIC_AGE, df = s)8 0.827702 0.268929 3.078 0.00209 **
ns(NUMERIC_AGE, df = s)9 -0.692540 0.151941 -4.558 5.20e-06 ***
PM_VISIT_LAST_2_YRS 0.325604 0.028814 11.300 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.276539 0.028921 9.562 < 2e-16 ***
AF_25K_GIFT 0.570474 0.043055 13.250 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.259873 0.006817 38.122 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.912117 0.042721 68.167 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.575610 0.025640 22.449 < 2e-16 ***
MG_250K_PLUS 1.075213 0.071617 15.013 < 2e-16 ***
Alumnus -0.565649 0.022023 -25.684 < 2e-16 ***
SEASON_TICKET_YEARS -0.022208 0.004460 -4.979 6.44e-07 ***
AFFINITY_SCORE 0.160207 0.006839 23.427 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.118849 0.022294 -5.331 9.88e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.208923 0.028639 7.295 3.11e-13 ***
MG_PR_MODEL_DESCTop Tier 0.589796 0.028307 20.836 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9651 on 18755 degrees of freedom
Multiple R-squared: 0.7257, Adjusted R-squared: 0.7254
F-statistic: 2256 on 22 and 18755 DF, p-value: < 2.2e-16
[[9]][[10]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3512 -0.6298 -0.1829 0.4973 6.3486
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.024470 0.118491 -0.207 0.83639
ns(NUMERIC_AGE, df = s)1 0.671216 0.106448 6.306 2.94e-10 ***
ns(NUMERIC_AGE, df = s)2 0.722151 0.137164 5.265 1.42e-07 ***
ns(NUMERIC_AGE, df = s)3 0.649514 0.125062 5.194 2.08e-07 ***
ns(NUMERIC_AGE, df = s)4 0.579473 0.119211 4.861 1.18e-06 ***
ns(NUMERIC_AGE, df = s)5 0.605008 0.125475 4.822 1.43e-06 ***
ns(NUMERIC_AGE, df = s)6 0.294908 0.124928 2.361 0.01825 *
ns(NUMERIC_AGE, df = s)7 -0.027618 0.092272 -0.299 0.76471
ns(NUMERIC_AGE, df = s)8 1.007782 0.266916 3.776 0.00016 ***
ns(NUMERIC_AGE, df = s)9 -0.678070 0.150608 -4.502 6.76e-06 ***
PM_VISIT_LAST_2_YRS 0.312174 0.028898 10.803 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.284044 0.029046 9.779 < 2e-16 ***
AF_25K_GIFT 0.613257 0.042915 14.290 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.255034 0.006821 37.391 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.918396 0.042687 68.368 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.582313 0.025695 22.662 < 2e-16 ***
MG_250K_PLUS 1.064423 0.071930 14.798 < 2e-16 ***
Alumnus -0.564290 0.022098 -25.536 < 2e-16 ***
SEASON_TICKET_YEARS -0.021537 0.004512 -4.773 1.83e-06 ***
AFFINITY_SCORE 0.161297 0.006852 23.541 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.111683 0.022347 -4.998 5.85e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.204245 0.028614 7.138 9.82e-13 ***
MG_PR_MODEL_DESCTop Tier 0.578858 0.028411 20.374 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.967 on 18751 degrees of freedom
Multiple R-squared: 0.7235, Adjusted R-squared: 0.7232
F-statistic: 2230 on 22 and 18751 DF, p-value: < 2.2e-16
[[10]]
[[10]][[1]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4338 -0.6232 -0.1776 0.4907 5.4696
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.045625 0.122081 0.374 0.708609
ns(NUMERIC_AGE, df = s)1 0.654440 0.110407 5.927 3.13e-09 ***
ns(NUMERIC_AGE, df = s)2 0.614194 0.139090 4.416 1.01e-05 ***
ns(NUMERIC_AGE, df = s)3 0.633304 0.128248 4.938 7.96e-07 ***
ns(NUMERIC_AGE, df = s)4 0.529191 0.127971 4.135 3.56e-05 ***
ns(NUMERIC_AGE, df = s)5 0.501455 0.127868 3.922 8.83e-05 ***
ns(NUMERIC_AGE, df = s)6 0.488045 0.133022 3.669 0.000244 ***
ns(NUMERIC_AGE, df = s)7 0.214799 0.128283 1.674 0.094066 .
ns(NUMERIC_AGE, df = s)8 -0.053982 0.094517 -0.571 0.567915
ns(NUMERIC_AGE, df = s)9 0.749495 0.275334 2.722 0.006492 **
ns(NUMERIC_AGE, df = s)10 -0.767586 0.158471 -4.844 1.28e-06 ***
PM_VISIT_LAST_2_YRS 0.311947 0.028737 10.855 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.274306 0.028874 9.500 < 2e-16 ***
AF_25K_GIFT 0.617249 0.042558 14.504 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.260948 0.006782 38.475 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.929075 0.042379 69.116 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.557944 0.025526 21.858 < 2e-16 ***
MG_250K_PLUS 1.127215 0.071603 15.743 < 2e-16 ***
Alumnus -0.569195 0.021887 -26.006 < 2e-16 ***
SEASON_TICKET_YEARS -0.019382 0.004460 -4.346 1.40e-05 ***
AFFINITY_SCORE 0.158130 0.006816 23.199 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.117494 0.022125 -5.310 1.11e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.215385 0.028345 7.599 3.13e-14 ***
MG_PR_MODEL_DESCTop Tier 0.596052 0.028195 21.140 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9607 on 18754 degrees of freedom
Multiple R-squared: 0.7273, Adjusted R-squared: 0.727
F-statistic: 2175 on 23 and 18754 DF, p-value: < 2.2e-16
[[10]][[2]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4781 -0.6215 -0.1742 0.4902 6.3395
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.065092 0.120479 0.540 0.58901
ns(NUMERIC_AGE, df = s)1 0.641767 0.108831 5.897 3.77e-09 ***
ns(NUMERIC_AGE, df = s)2 0.555837 0.137488 4.043 5.30e-05 ***
ns(NUMERIC_AGE, df = s)3 0.656240 0.126805 5.175 2.30e-07 ***
ns(NUMERIC_AGE, df = s)4 0.505317 0.126532 3.994 6.53e-05 ***
ns(NUMERIC_AGE, df = s)5 0.501402 0.126368 3.968 7.28e-05 ***
ns(NUMERIC_AGE, df = s)6 0.426201 0.131794 3.234 0.00122 **
ns(NUMERIC_AGE, df = s)7 0.218974 0.126854 1.726 0.08433 .
ns(NUMERIC_AGE, df = s)8 -0.099782 0.092097 -1.083 0.27862
ns(NUMERIC_AGE, df = s)9 0.787760 0.270396 2.913 0.00358 **
ns(NUMERIC_AGE, df = s)10 -0.619381 0.142122 -4.358 1.32e-05 ***
PM_VISIT_LAST_2_YRS 0.302919 0.028789 10.522 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.279885 0.028953 9.667 < 2e-16 ***
AF_25K_GIFT 0.589053 0.043186 13.640 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.259871 0.006769 38.392 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.912623 0.042390 68.710 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.557267 0.025565 21.798 < 2e-16 ***
MG_250K_PLUS 1.166267 0.070037 16.652 < 2e-16 ***
Alumnus -0.564245 0.021954 -25.701 < 2e-16 ***
SEASON_TICKET_YEARS -0.018816 0.004504 -4.177 2.96e-05 ***
AFFINITY_SCORE 0.158633 0.006813 23.285 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.122624 0.022106 -5.547 2.94e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.208872 0.028343 7.370 1.78e-13 ***
MG_PR_MODEL_DESCTop Tier 0.601409 0.028341 21.220 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9606 on 18754 degrees of freedom
Multiple R-squared: 0.7276, Adjusted R-squared: 0.7273
F-statistic: 2178 on 23 and 18754 DF, p-value: < 2.2e-16
[[10]][[3]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4021 -0.6237 -0.1802 0.4930 6.3664
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.119698 0.124563 0.961 0.336594
ns(NUMERIC_AGE, df = s)1 0.577046 0.112454 5.131 2.90e-07 ***
ns(NUMERIC_AGE, df = s)2 0.517996 0.141867 3.651 0.000262 ***
ns(NUMERIC_AGE, df = s)3 0.572305 0.130679 4.379 1.20e-05 ***
ns(NUMERIC_AGE, df = s)4 0.459434 0.130582 3.518 0.000435 ***
ns(NUMERIC_AGE, df = s)5 0.405757 0.130265 3.115 0.001843 **
ns(NUMERIC_AGE, df = s)6 0.389882 0.135504 2.877 0.004016 **
ns(NUMERIC_AGE, df = s)7 0.132434 0.130660 1.014 0.310798
ns(NUMERIC_AGE, df = s)8 -0.137503 0.095122 -1.446 0.148323
ns(NUMERIC_AGE, df = s)9 0.645110 0.280645 2.299 0.021535 *
ns(NUMERIC_AGE, df = s)10 -0.681722 0.157370 -4.332 1.49e-05 ***
PM_VISIT_LAST_2_YRS 0.320753 0.028798 11.138 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.263349 0.028886 9.117 < 2e-16 ***
AF_25K_GIFT 0.614628 0.042750 14.377 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.262958 0.006795 38.698 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.915344 0.042494 68.605 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.563173 0.025666 21.943 < 2e-16 ***
MG_250K_PLUS 1.090582 0.071111 15.336 < 2e-16 ***
Alumnus -0.571488 0.021946 -26.040 < 2e-16 ***
SEASON_TICKET_YEARS -0.020177 0.004419 -4.566 5.00e-06 ***
AFFINITY_SCORE 0.160404 0.006843 23.442 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.110160 0.022175 -4.968 6.83e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.207948 0.028457 7.307 2.83e-13 ***
MG_PR_MODEL_DESCTop Tier 0.595617 0.028350 21.010 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9637 on 18754 degrees of freedom
Multiple R-squared: 0.7262, Adjusted R-squared: 0.7259
F-statistic: 2163 on 23 and 18754 DF, p-value: < 2.2e-16
[[10]][[4]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4075 -0.6248 -0.1795 0.4937 6.3254
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.037467 0.121563 0.308 0.757924
ns(NUMERIC_AGE, df = s)1 0.671386 0.109812 6.114 9.91e-10 ***
ns(NUMERIC_AGE, df = s)2 0.629826 0.138650 4.543 5.59e-06 ***
ns(NUMERIC_AGE, df = s)3 0.650725 0.127889 5.088 3.65e-07 ***
ns(NUMERIC_AGE, df = s)4 0.573117 0.127570 4.493 7.08e-06 ***
ns(NUMERIC_AGE, df = s)5 0.500060 0.127306 3.928 8.60e-05 ***
ns(NUMERIC_AGE, df = s)6 0.500646 0.132678 3.773 0.000162 ***
ns(NUMERIC_AGE, df = s)7 0.238443 0.127813 1.866 0.062118 .
ns(NUMERIC_AGE, df = s)8 -0.093824 0.094576 -0.992 0.321188
ns(NUMERIC_AGE, df = s)9 0.902970 0.274344 3.291 0.000999 ***
ns(NUMERIC_AGE, df = s)10 -0.591629 0.159489 -3.710 0.000208 ***
PM_VISIT_LAST_2_YRS 0.319836 0.028959 11.045 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.288517 0.028920 9.976 < 2e-16 ***
AF_25K_GIFT 0.626357 0.042886 14.605 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.259593 0.006803 38.160 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.924222 0.042453 68.881 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.577937 0.025558 22.612 < 2e-16 ***
MG_250K_PLUS 1.093082 0.071374 15.315 < 2e-16 ***
Alumnus -0.568680 0.021961 -25.895 < 2e-16 ***
SEASON_TICKET_YEARS -0.019762 0.004442 -4.449 8.66e-06 ***
AFFINITY_SCORE 0.155868 0.006849 22.759 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.123690 0.022221 -5.566 2.64e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.193754 0.028530 6.791 1.15e-11 ***
MG_PR_MODEL_DESCTop Tier 0.577720 0.028384 20.354 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9644 on 18754 degrees of freedom
Multiple R-squared: 0.7247, Adjusted R-squared: 0.7244
F-statistic: 2146 on 23 and 18754 DF, p-value: < 2.2e-16
[[10]][[5]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.0884 -0.6274 -0.1803 0.4969 6.3486
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.087512 0.123370 0.709 0.478118
ns(NUMERIC_AGE, df = s)1 0.616635 0.111417 5.534 3.16e-08 ***
ns(NUMERIC_AGE, df = s)2 0.569451 0.140751 4.046 5.24e-05 ***
ns(NUMERIC_AGE, df = s)3 0.602023 0.129630 4.644 3.44e-06 ***
ns(NUMERIC_AGE, df = s)4 0.502361 0.129488 3.880 0.000105 ***
ns(NUMERIC_AGE, df = s)5 0.460388 0.129285 3.561 0.000370 ***
ns(NUMERIC_AGE, df = s)6 0.441658 0.134451 3.285 0.001022 **
ns(NUMERIC_AGE, df = s)7 0.166120 0.129611 1.282 0.199969
ns(NUMERIC_AGE, df = s)8 -0.063914 0.095323 -0.670 0.502551
ns(NUMERIC_AGE, df = s)9 0.719871 0.278743 2.583 0.009814 **
ns(NUMERIC_AGE, df = s)10 -0.711998 0.161236 -4.416 1.01e-05 ***
PM_VISIT_LAST_2_YRS 0.339703 0.029087 11.679 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.277797 0.029130 9.537 < 2e-16 ***
AF_25K_GIFT 0.608039 0.043043 14.126 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.258764 0.006808 38.009 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.916552 0.042668 68.355 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.558397 0.025751 21.685 < 2e-16 ***
MG_250K_PLUS 1.098807 0.071863 15.290 < 2e-16 ***
Alumnus -0.569354 0.022062 -25.807 < 2e-16 ***
SEASON_TICKET_YEARS -0.020832 0.004514 -4.615 3.96e-06 ***
AFFINITY_SCORE 0.158742 0.006865 23.123 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.116212 0.022300 -5.211 1.89e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.221227 0.028522 7.756 9.19e-15 ***
MG_PR_MODEL_DESCTop Tier 0.595238 0.028412 20.950 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9678 on 18754 degrees of freedom
Multiple R-squared: 0.7233, Adjusted R-squared: 0.7229
F-statistic: 2131 on 23 and 18754 DF, p-value: < 2.2e-16
[[10]][[6]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4442 -0.6201 -0.1804 0.4929 6.3621
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.026568 0.121043 0.219 0.826273
ns(NUMERIC_AGE, df = s)1 0.690341 0.109284 6.317 2.73e-10 ***
ns(NUMERIC_AGE, df = s)2 0.684220 0.138333 4.946 7.63e-07 ***
ns(NUMERIC_AGE, df = s)3 0.685649 0.127415 5.381 7.49e-08 ***
ns(NUMERIC_AGE, df = s)4 0.586344 0.127087 4.614 3.98e-06 ***
ns(NUMERIC_AGE, df = s)5 0.526686 0.126893 4.151 3.33e-05 ***
ns(NUMERIC_AGE, df = s)6 0.504157 0.132143 3.815 0.000136 ***
ns(NUMERIC_AGE, df = s)7 0.250931 0.127334 1.971 0.048779 *
ns(NUMERIC_AGE, df = s)8 -0.059144 0.094519 -0.626 0.531498
ns(NUMERIC_AGE, df = s)9 0.883715 0.273672 3.229 0.001244 **
ns(NUMERIC_AGE, df = s)10 -0.694774 0.160614 -4.326 1.53e-05 ***
PM_VISIT_LAST_2_YRS 0.284618 0.029130 9.771 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.301587 0.028878 10.443 < 2e-16 ***
AF_25K_GIFT 0.615356 0.043467 14.157 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.260368 0.006794 38.325 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.912634 0.042514 68.510 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.561952 0.025569 21.978 < 2e-16 ***
MG_250K_PLUS 1.142162 0.072224 15.814 < 2e-16 ***
Alumnus -0.577556 0.021868 -26.412 < 2e-16 ***
SEASON_TICKET_YEARS -0.019100 0.004441 -4.301 1.71e-05 ***
AFFINITY_SCORE 0.158630 0.006840 23.193 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.131752 0.022195 -5.936 2.97e-09 ***
MG_PR_MODEL_DESCMiddle Tier 0.192383 0.028438 6.765 1.37e-11 ***
MG_PR_MODEL_DESCTop Tier 0.567270 0.028285 20.056 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9621 on 18754 degrees of freedom
Multiple R-squared: 0.7258, Adjusted R-squared: 0.7254
F-statistic: 2158 on 23 and 18754 DF, p-value: < 2.2e-16
[[10]][[7]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4050 -0.6253 -0.1823 0.4948 6.3325
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.027658 0.120547 0.229 0.818531
ns(NUMERIC_AGE, df = s)1 0.645226 0.108676 5.937 2.95e-09 ***
ns(NUMERIC_AGE, df = s)2 0.640236 0.137537 4.655 3.26e-06 ***
ns(NUMERIC_AGE, df = s)3 0.657866 0.126826 5.187 2.16e-07 ***
ns(NUMERIC_AGE, df = s)4 0.530040 0.126578 4.187 2.83e-05 ***
ns(NUMERIC_AGE, df = s)5 0.517861 0.126416 4.096 4.21e-05 ***
ns(NUMERIC_AGE, df = s)6 0.491601 0.131849 3.729 0.000193 ***
ns(NUMERIC_AGE, df = s)7 0.209767 0.126854 1.654 0.098224 .
ns(NUMERIC_AGE, df = s)8 -0.106291 0.094396 -1.126 0.260174
ns(NUMERIC_AGE, df = s)9 0.895662 0.272846 3.283 0.001030 **
ns(NUMERIC_AGE, df = s)10 -0.616409 0.163155 -3.778 0.000159 ***
PM_VISIT_LAST_2_YRS 0.306975 0.028935 10.609 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.299437 0.028889 10.365 < 2e-16 ***
AF_25K_GIFT 0.623662 0.042915 14.532 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.263054 0.006790 38.739 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.899153 0.042511 68.198 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.567405 0.025617 22.149 < 2e-16 ***
MG_250K_PLUS 1.095521 0.072595 15.091 < 2e-16 ***
Alumnus -0.569044 0.021947 -25.928 < 2e-16 ***
SEASON_TICKET_YEARS -0.019062 0.004531 -4.207 2.60e-05 ***
AFFINITY_SCORE 0.157780 0.006843 23.056 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.108744 0.022214 -4.895 9.89e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.213414 0.028471 7.496 6.88e-14 ***
MG_PR_MODEL_DESCTop Tier 0.580368 0.028332 20.484 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9633 on 18754 degrees of freedom
Multiple R-squared: 0.7251, Adjusted R-squared: 0.7248
F-statistic: 2151 on 23 and 18754 DF, p-value: < 2.2e-16
[[10]][[8]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.4532 -0.6294 -0.1756 0.4950 6.3330
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.037361 0.123113 0.303 0.7615
ns(NUMERIC_AGE, df = s)1 0.675219 0.111148 6.075 1.26e-09 ***
ns(NUMERIC_AGE, df = s)2 0.612503 0.140335 4.365 1.28e-05 ***
ns(NUMERIC_AGE, df = s)3 0.658857 0.129439 5.090 3.61e-07 ***
ns(NUMERIC_AGE, df = s)4 0.554129 0.129296 4.286 1.83e-05 ***
ns(NUMERIC_AGE, df = s)5 0.513087 0.129042 3.976 7.03e-05 ***
ns(NUMERIC_AGE, df = s)6 0.526015 0.134296 3.917 9.00e-05 ***
ns(NUMERIC_AGE, df = s)7 0.202513 0.129352 1.566 0.1175
ns(NUMERIC_AGE, df = s)8 -0.034019 0.094975 -0.358 0.7202
ns(NUMERIC_AGE, df = s)9 0.827350 0.277806 2.978 0.0029 **
ns(NUMERIC_AGE, df = s)10 -0.671168 0.158346 -4.239 2.26e-05 ***
PM_VISIT_LAST_2_YRS 0.319250 0.029183 10.940 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.287108 0.029093 9.869 < 2e-16 ***
AF_25K_GIFT 0.577569 0.043674 13.224 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.258790 0.006842 37.823 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.901147 0.042609 68.087 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.555450 0.025783 21.543 < 2e-16 ***
MG_250K_PLUS 1.128778 0.072288 15.615 < 2e-16 ***
Alumnus -0.564653 0.022088 -25.564 < 2e-16 ***
SEASON_TICKET_YEARS -0.018866 0.004483 -4.208 2.59e-05 ***
AFFINITY_SCORE 0.159796 0.006873 23.249 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.117215 0.022256 -5.267 1.40e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.200730 0.028548 7.031 2.12e-12 ***
MG_PR_MODEL_DESCTop Tier 0.586308 0.028417 20.632 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9678 on 18754 degrees of freedom
Multiple R-squared: 0.7221, Adjusted R-squared: 0.7218
F-statistic: 2119 on 23 and 18754 DF, p-value: < 2.2e-16
[[10]][[9]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3955 -0.6228 -0.1819 0.4895 6.3636
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.038515 0.121883 0.316 0.752009
ns(NUMERIC_AGE, df = s)1 0.662175 0.109797 6.031 1.66e-09 ***
ns(NUMERIC_AGE, df = s)2 0.601090 0.139208 4.318 1.58e-05 ***
ns(NUMERIC_AGE, df = s)3 0.628427 0.128174 4.903 9.52e-07 ***
ns(NUMERIC_AGE, df = s)4 0.539485 0.127915 4.218 2.48e-05 ***
ns(NUMERIC_AGE, df = s)5 0.509153 0.127714 3.987 6.73e-05 ***
ns(NUMERIC_AGE, df = s)6 0.469417 0.133108 3.527 0.000422 ***
ns(NUMERIC_AGE, df = s)7 0.209926 0.128019 1.640 0.101064
ns(NUMERIC_AGE, df = s)8 -0.037962 0.094769 -0.401 0.688738
ns(NUMERIC_AGE, df = s)9 0.774043 0.275217 2.812 0.004921 **
ns(NUMERIC_AGE, df = s)10 -0.700509 0.159991 -4.378 1.20e-05 ***
PM_VISIT_LAST_2_YRS 0.325519 0.028820 11.295 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.276466 0.028923 9.559 < 2e-16 ***
AF_25K_GIFT 0.570900 0.043057 13.259 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.259924 0.006816 38.134 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.911847 0.042717 68.165 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.575435 0.025637 22.445 < 2e-16 ***
MG_250K_PLUS 1.074981 0.071624 15.009 < 2e-16 ***
Alumnus -0.564507 0.021966 -25.699 < 2e-16 ***
SEASON_TICKET_YEARS -0.022181 0.004460 -4.973 6.65e-07 ***
AFFINITY_SCORE 0.160304 0.006837 23.448 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.118835 0.022296 -5.330 9.94e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.208983 0.028641 7.297 3.07e-13 ***
MG_PR_MODEL_DESCTop Tier 0.589832 0.028309 20.836 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9651 on 18754 degrees of freedom
Multiple R-squared: 0.7257, Adjusted R-squared: 0.7254
F-statistic: 2158 on 23 and 18754 DF, p-value: < 2.2e-16
[[10]][[10]]
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = s) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-4.3524 -0.6300 -0.1813 0.4976 6.3522
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.020997 0.120858 -0.174 0.862079
ns(NUMERIC_AGE, df = s)1 0.695485 0.109060 6.377 1.85e-10 ***
ns(NUMERIC_AGE, df = s)2 0.689322 0.137927 4.998 5.85e-07 ***
ns(NUMERIC_AGE, df = s)3 0.684533 0.126741 5.401 6.71e-08 ***
ns(NUMERIC_AGE, df = s)4 0.600084 0.124075 4.836 1.33e-06 ***
ns(NUMERIC_AGE, df = s)5 0.558748 0.129208 4.324 1.54e-05 ***
ns(NUMERIC_AGE, df = s)6 0.543331 0.132323 4.106 4.04e-05 ***
ns(NUMERIC_AGE, df = s)7 0.277227 0.126814 2.186 0.028821 *
ns(NUMERIC_AGE, df = s)8 -0.023697 0.095017 -0.249 0.803053
ns(NUMERIC_AGE, df = s)9 0.965485 0.272955 3.537 0.000405 ***
ns(NUMERIC_AGE, df = s)10 -0.695447 0.159381 -4.363 1.29e-05 ***
PM_VISIT_LAST_2_YRS 0.311908 0.028905 10.791 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.284065 0.029048 9.779 < 2e-16 ***
AF_25K_GIFT 0.613754 0.042917 14.301 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.255098 0.006820 37.404 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.917974 0.042682 68.365 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.582031 0.025696 22.651 < 2e-16 ***
MG_250K_PLUS 1.064365 0.071938 14.796 < 2e-16 ***
Alumnus -0.562790 0.022052 -25.521 < 2e-16 ***
SEASON_TICKET_YEARS -0.021521 0.004512 -4.769 1.86e-06 ***
AFFINITY_SCORE 0.161418 0.006851 23.560 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.111784 0.022350 -5.002 5.74e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.204190 0.028616 7.136 9.99e-13 ***
MG_PR_MODEL_DESCTop Tier 0.578817 0.028414 20.371 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9671 on 18750 degrees of freedom
Multiple R-squared: 0.7235, Adjusted R-squared: 0.7231
F-statistic: 2133 on 23 and 18750 DF, p-value: < 2.2e-16Campaign final models
summary(clm_final)
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ACTIVE_PROPOSALS +
AGE + PM_VISIT_LAST_2_YRS + VISITS_5PLUS + AF_25K_GIFT +
GAVE_IN_LAST_3_YRS + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DEEP_ENGAGEMENT + CHICAGO_HOME, data = mdat %>%
filter(rownum %in% unlist(xval_inds)))
Residuals:
Min 1Q Median 3Q Max
-4.8516 -1.2287 -0.1483 0.9862 5.6918
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.34421 0.02665 50.430 < 2e-16 ***
ACTIVE_PROPOSALS 0.28318 0.04351 6.509 7.75e-11 ***
AGE -0.22153 0.02035 -10.886 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.61254 0.04596 13.329 < 2e-16 ***
VISITS_5PLUS 0.74935 0.03304 22.677 < 2e-16 ***
AF_25K_GIFT 0.83750 0.05752 14.561 < 2e-16 ***
GAVE_IN_LAST_3_YRS 2.04013 0.02217 92.019 < 2e-16 ***
MG_250K_PLUS 1.10852 0.09647 11.491 < 2e-16 ***
PRESIDENT_VISIT 0.45473 0.08038 5.657 1.56e-08 ***
TRUSTEE_OR_ADVISORY_BOARD 0.27001 0.04310 6.265 3.81e-10 ***
Alumnus -0.11548 0.02628 -4.395 1.11e-05 ***
DEEP_ENGAGEMENT 0.33019 0.02181 15.137 < 2e-16 ***
CHICAGO_HOME 0.09710 0.02119 4.582 4.63e-06 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 1.344 on 20851 degrees of freedom
Multiple R-squared: 0.4654, Adjusted R-squared: 0.4651
F-statistic: 1512 on 12 and 20851 DF, p-value: < 2.2e-16summary(clmap_final)
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = 5) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = mdat %>% filter(rownum %in%
unlist(xval_inds)))
Residuals:
Min 1Q Median 3Q Max
-4.4145 -0.6252 -0.1852 0.4953 6.2824
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.273983 0.089046 3.077 0.00209 **
ns(NUMERIC_AGE, df = 5)1 0.372461 0.081347 4.579 4.71e-06 ***
ns(NUMERIC_AGE, df = 5)2 0.236252 0.095335 2.478 0.01322 *
ns(NUMERIC_AGE, df = 5)3 -0.385478 0.065826 -5.856 4.81e-09 ***
ns(NUMERIC_AGE, df = 5)4 0.472451 0.201811 2.341 0.01924 *
ns(NUMERIC_AGE, df = 5)5 -0.466161 0.108905 -4.280 1.87e-05 ***
PM_VISIT_LAST_2_YRS 0.313757 0.027446 11.432 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.282839 0.027478 10.293 < 2e-16 ***
AF_25K_GIFT 0.604838 0.040840 14.810 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.259760 0.006446 40.295 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.915133 0.040261 72.406 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.566151 0.024289 23.309 < 2e-16 ***
MG_250K_PLUS 1.106584 0.067980 16.278 < 2e-16 ***
Alumnus -0.572008 0.020363 -28.091 < 2e-16 ***
SEASON_TICKET_YEARS -0.019864 0.004247 -4.677 2.93e-06 ***
AFFINITY_SCORE 0.158965 0.006473 24.557 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.118435 0.021085 -5.617 1.97e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.206051 0.027034 7.622 2.60e-14 ***
MG_PR_MODEL_DESCTop Tier 0.585706 0.026890 21.781 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9646 on 20845 degrees of freedom
Multiple R-squared: 0.7248, Adjusted R-squared: 0.7246
F-statistic: 3050 on 18 and 20845 DF, p-value: < 2.2e-16Transaction cultivation results
lapply(tlms, function(x) summary(x))[[1]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ACTIVE_PROPOSALS +
AGE + PM_VISIT_LAST_2_YRS + VISITS_5PLUS + AF_25K_GIFT +
GAVE_IN_LAST_3_YRS + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DEEP_ENGAGEMENT + CHICAGO_HOME, data = .)
Residuals:
Min 1Q Median 3Q Max
-3.3186 -0.5623 0.1469 0.7486 3.8462
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.76745 0.02333 75.755 < 2e-16 ***
ACTIVE_PROPOSALS 0.23081 0.03785 6.098 1.10e-09 ***
AGE 0.28694 0.01779 16.128 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.25410 0.04006 6.344 2.30e-10 ***
VISITS_5PLUS 0.73105 0.02894 25.258 < 2e-16 ***
AF_25K_GIFT 0.87487 0.04999 17.502 < 2e-16 ***
GAVE_IN_LAST_3_YRS 0.77836 0.01936 40.198 < 2e-16 ***
MG_250K_PLUS 1.78288 0.08481 21.023 < 2e-16 ***
PRESIDENT_VISIT 0.14916 0.07040 2.119 0.0341 *
TRUSTEE_OR_ADVISORY_BOARD 0.34185 0.03747 9.123 < 2e-16 ***
Alumnus 0.02082 0.02301 0.905 0.3657
DEEP_ENGAGEMENT 0.18227 0.01910 9.542 < 2e-16 ***
CHICAGO_HOME 0.13213 0.01853 7.132 1.02e-12 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 1.116 on 18765 degrees of freedom
Multiple R-squared: 0.3308, Adjusted R-squared: 0.3304
F-statistic: 773.1 on 12 and 18765 DF, p-value: < 2.2e-16
[[2]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ACTIVE_PROPOSALS +
AGE + PM_VISIT_LAST_2_YRS + VISITS_5PLUS + AF_25K_GIFT +
GAVE_IN_LAST_3_YRS + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DEEP_ENGAGEMENT + CHICAGO_HOME, data = .)
Residuals:
Min 1Q Median 3Q Max
-3.3375 -0.5622 0.1450 0.7466 3.8217
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.75663 0.02338 75.147 < 2e-16 ***
ACTIVE_PROPOSALS 0.24415 0.03814 6.402 1.57e-10 ***
AGE 0.28706 0.01783 16.102 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.25264 0.04025 6.276 3.54e-10 ***
VISITS_5PLUS 0.72456 0.02901 24.977 < 2e-16 ***
AF_25K_GIFT 0.82841 0.05085 16.292 < 2e-16 ***
GAVE_IN_LAST_3_YRS 0.78491 0.01940 40.460 < 2e-16 ***
MG_250K_PLUS 1.79186 0.08278 21.645 < 2e-16 ***
PRESIDENT_VISIT 0.17113 0.07032 2.434 0.015 *
TRUSTEE_OR_ADVISORY_BOARD 0.35174 0.03777 9.313 < 2e-16 ***
Alumnus 0.02498 0.02306 1.083 0.279
DEEP_ENGAGEMENT 0.18836 0.01908 9.872 < 2e-16 ***
CHICAGO_HOME 0.12592 0.01856 6.784 1.21e-11 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 1.117 on 18765 degrees of freedom
Multiple R-squared: 0.3327, Adjusted R-squared: 0.3323
F-statistic: 779.8 on 12 and 18765 DF, p-value: < 2.2e-16
[[3]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ACTIVE_PROPOSALS +
AGE + PM_VISIT_LAST_2_YRS + VISITS_5PLUS + AF_25K_GIFT +
GAVE_IN_LAST_3_YRS + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DEEP_ENGAGEMENT + CHICAGO_HOME, data = .)
Residuals:
Min 1Q Median 3Q Max
-3.3315 -0.5628 0.1504 0.7505 3.8532
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.75160 0.02334 75.043 < 2e-16 ***
ACTIVE_PROPOSALS 0.22904 0.03821 5.994 2.08e-09 ***
AGE 0.28566 0.01789 15.970 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.27140 0.04037 6.723 1.83e-11 ***
VISITS_5PLUS 0.72201 0.02898 24.910 < 2e-16 ***
AF_25K_GIFT 0.85276 0.05024 16.973 < 2e-16 ***
GAVE_IN_LAST_3_YRS 0.78434 0.01948 40.258 < 2e-16 ***
MG_250K_PLUS 1.75443 0.08418 20.841 < 2e-16 ***
PRESIDENT_VISIT 0.15976 0.07009 2.279 0.0227 *
TRUSTEE_OR_ADVISORY_BOARD 0.36026 0.03794 9.495 < 2e-16 ***
Alumnus 0.02844 0.02303 1.235 0.2167
DEEP_ENGAGEMENT 0.19772 0.01912 10.340 < 2e-16 ***
CHICAGO_HOME 0.12891 0.01864 6.915 4.82e-12 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 1.12 on 18765 degrees of freedom
Multiple R-squared: 0.3324, Adjusted R-squared: 0.332
F-statistic: 778.8 on 12 and 18765 DF, p-value: < 2.2e-16
[[4]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ACTIVE_PROPOSALS +
AGE + PM_VISIT_LAST_2_YRS + VISITS_5PLUS + AF_25K_GIFT +
GAVE_IN_LAST_3_YRS + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DEEP_ENGAGEMENT + CHICAGO_HOME, data = .)
Residuals:
Min 1Q Median 3Q Max
-3.3606 -0.5623 0.1493 0.7406 3.8029
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.778531 0.023248 76.502 < 2e-16 ***
ACTIVE_PROPOSALS 0.214159 0.038120 5.618 1.96e-08 ***
AGE 0.284538 0.017781 16.002 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.271383 0.040322 6.730 1.74e-11 ***
VISITS_5PLUS 0.717755 0.028830 24.896 < 2e-16 ***
AF_25K_GIFT 0.878243 0.050065 17.542 < 2e-16 ***
GAVE_IN_LAST_3_YRS 0.779877 0.019349 40.305 < 2e-16 ***
MG_250K_PLUS 1.720435 0.084170 20.440 < 2e-16 ***
PRESIDENT_VISIT 0.198072 0.070362 2.815 0.00488 **
TRUSTEE_OR_ADVISORY_BOARD 0.364242 0.037852 9.623 < 2e-16 ***
Alumnus 0.008213 0.022931 0.358 0.72022
DEEP_ENGAGEMENT 0.180692 0.019013 9.504 < 2e-16 ***
CHICAGO_HOME 0.128785 0.018510 6.958 3.57e-12 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 1.115 on 18765 degrees of freedom
Multiple R-squared: 0.3309, Adjusted R-squared: 0.3305
F-statistic: 773.3 on 12 and 18765 DF, p-value: < 2.2e-16
[[5]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ACTIVE_PROPOSALS +
AGE + PM_VISIT_LAST_2_YRS + VISITS_5PLUS + AF_25K_GIFT +
GAVE_IN_LAST_3_YRS + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DEEP_ENGAGEMENT + CHICAGO_HOME, data = .)
Residuals:
Min 1Q Median 3Q Max
-3.3484 -0.5621 0.1438 0.7444 3.8409
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.76082 0.02340 75.241 < 2e-16 ***
ACTIVE_PROPOSALS 0.22065 0.03842 5.744 9.41e-09 ***
AGE 0.27916 0.01787 15.621 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.27569 0.04043 6.820 9.40e-12 ***
VISITS_5PLUS 0.72535 0.02915 24.886 < 2e-16 ***
AF_25K_GIFT 0.87277 0.05036 17.332 < 2e-16 ***
GAVE_IN_LAST_3_YRS 0.78059 0.01948 40.077 < 2e-16 ***
MG_250K_PLUS 1.72772 0.08436 20.481 < 2e-16 ***
PRESIDENT_VISIT 0.17132 0.06995 2.449 0.0143 *
TRUSTEE_OR_ADVISORY_BOARD 0.35803 0.03773 9.490 < 2e-16 ***
Alumnus 0.02431 0.02308 1.053 0.2923
DEEP_ENGAGEMENT 0.19459 0.01917 10.152 < 2e-16 ***
CHICAGO_HOME 0.12749 0.01859 6.858 7.22e-12 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 1.119 on 18765 degrees of freedom
Multiple R-squared: 0.3294, Adjusted R-squared: 0.329
F-statistic: 768.2 on 12 and 18765 DF, p-value: < 2.2e-16
[[6]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ACTIVE_PROPOSALS +
AGE + PM_VISIT_LAST_2_YRS + VISITS_5PLUS + AF_25K_GIFT +
GAVE_IN_LAST_3_YRS + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DEEP_ENGAGEMENT + CHICAGO_HOME, data = .)
Residuals:
Min 1Q Median 3Q Max
-3.3704 -0.5681 0.1468 0.7489 3.7989
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.76913 0.02337 75.697 < 2e-16 ***
ACTIVE_PROPOSALS 0.22722 0.03831 5.931 3.06e-09 ***
AGE 0.28119 0.01784 15.763 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.26562 0.04057 6.547 6.02e-11 ***
VISITS_5PLUS 0.72319 0.02900 24.939 < 2e-16 ***
AF_25K_GIFT 0.84522 0.05116 16.522 < 2e-16 ***
GAVE_IN_LAST_3_YRS 0.79322 0.01946 40.755 < 2e-16 ***
MG_250K_PLUS 1.76585 0.08572 20.599 < 2e-16 ***
PRESIDENT_VISIT 0.19841 0.07114 2.789 0.00529 **
TRUSTEE_OR_ADVISORY_BOARD 0.35323 0.03775 9.356 < 2e-16 ***
Alumnus 0.01008 0.02300 0.438 0.66114
DEEP_ENGAGEMENT 0.18683 0.01912 9.770 < 2e-16 ***
CHICAGO_HOME 0.13176 0.01855 7.101 1.28e-12 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 1.119 on 18765 degrees of freedom
Multiple R-squared: 0.3287, Adjusted R-squared: 0.3283
F-statistic: 765.9 on 12 and 18765 DF, p-value: < 2.2e-16
[[7]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ACTIVE_PROPOSALS +
AGE + PM_VISIT_LAST_2_YRS + VISITS_5PLUS + AF_25K_GIFT +
GAVE_IN_LAST_3_YRS + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DEEP_ENGAGEMENT + CHICAGO_HOME, data = .)
Residuals:
Min 1Q Median 3Q Max
-3.2998 -0.5609 0.1453 0.7529 3.7220
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.75324 0.02349 74.653 < 2e-16 ***
ACTIVE_PROPOSALS 0.18994 0.03832 4.957 7.22e-07 ***
AGE 0.28183 0.01787 15.769 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.27612 0.04031 6.850 7.61e-12 ***
VISITS_5PLUS 0.72514 0.02896 25.039 < 2e-16 ***
AF_25K_GIFT 0.87361 0.05028 17.373 < 2e-16 ***
GAVE_IN_LAST_3_YRS 0.78174 0.01950 40.095 < 2e-16 ***
MG_250K_PLUS 1.80905 0.08569 21.112 < 2e-16 ***
PRESIDENT_VISIT 0.21417 0.07052 3.037 0.00239 **
TRUSTEE_OR_ADVISORY_BOARD 0.36563 0.03760 9.723 < 2e-16 ***
Alumnus 0.03023 0.02311 1.308 0.19086
DEEP_ENGAGEMENT 0.18909 0.01923 9.835 < 2e-16 ***
CHICAGO_HOME 0.13609 0.01859 7.319 2.60e-13 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 1.12 on 18765 degrees of freedom
Multiple R-squared: 0.3282, Adjusted R-squared: 0.3278
F-statistic: 764.1 on 12 and 18765 DF, p-value: < 2.2e-16
[[8]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ACTIVE_PROPOSALS +
AGE + PM_VISIT_LAST_2_YRS + VISITS_5PLUS + AF_25K_GIFT +
GAVE_IN_LAST_3_YRS + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DEEP_ENGAGEMENT + CHICAGO_HOME, data = .)
Residuals:
Min 1Q Median 3Q Max
-3.3759 -0.5612 0.1379 0.7464 3.8046
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.75493 0.02335 75.166 < 2e-16 ***
ACTIVE_PROPOSALS 0.22829 0.03810 5.992 2.12e-09 ***
AGE 0.29702 0.01781 16.681 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.27159 0.04042 6.719 1.88e-11 ***
VISITS_5PLUS 0.72938 0.02894 25.204 < 2e-16 ***
AF_25K_GIFT 0.84439 0.05070 16.653 < 2e-16 ***
GAVE_IN_LAST_3_YRS 0.78387 0.01941 40.395 < 2e-16 ***
MG_250K_PLUS 1.76296 0.08481 20.787 < 2e-16 ***
PRESIDENT_VISIT 0.20579 0.07095 2.901 0.00373 **
TRUSTEE_OR_ADVISORY_BOARD 0.34503 0.03814 9.046 < 2e-16 ***
Alumnus 0.02668 0.02302 1.159 0.24644
DEEP_ENGAGEMENT 0.18590 0.01908 9.740 < 2e-16 ***
CHICAGO_HOME 0.13296 0.01858 7.158 8.49e-13 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 1.116 on 18765 degrees of freedom
Multiple R-squared: 0.3305, Adjusted R-squared: 0.3301
F-statistic: 772 on 12 and 18765 DF, p-value: < 2.2e-16
[[9]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ACTIVE_PROPOSALS +
AGE + PM_VISIT_LAST_2_YRS + VISITS_5PLUS + AF_25K_GIFT +
GAVE_IN_LAST_3_YRS + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DEEP_ENGAGEMENT + CHICAGO_HOME, data = .)
Residuals:
Min 1Q Median 3Q Max
-3.3025 -0.5698 0.1384 0.7503 3.8497
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.76942 0.02332 75.860 < 2e-16 ***
ACTIVE_PROPOSALS 0.22474 0.03779 5.947 2.78e-09 ***
AGE 0.28791 0.01787 16.116 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.25554 0.03994 6.398 1.62e-10 ***
VISITS_5PLUS 0.72293 0.02900 24.930 < 2e-16 ***
AF_25K_GIFT 0.84878 0.05042 16.833 < 2e-16 ***
GAVE_IN_LAST_3_YRS 0.78918 0.01947 40.543 < 2e-16 ***
MG_250K_PLUS 1.75295 0.08456 20.730 < 2e-16 ***
PRESIDENT_VISIT 0.14977 0.07013 2.136 0.0327 *
TRUSTEE_OR_ADVISORY_BOARD 0.35549 0.03774 9.420 < 2e-16 ***
Alumnus 0.01547 0.02299 0.673 0.5009
DEEP_ENGAGEMENT 0.18007 0.01914 9.408 < 2e-16 ***
CHICAGO_HOME 0.13009 0.01857 7.006 2.53e-12 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 1.118 on 18765 degrees of freedom
Multiple R-squared: 0.331, Adjusted R-squared: 0.3306
F-statistic: 773.6 on 12 and 18765 DF, p-value: < 2.2e-16
[[10]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ACTIVE_PROPOSALS +
AGE + PM_VISIT_LAST_2_YRS + VISITS_5PLUS + AF_25K_GIFT +
GAVE_IN_LAST_3_YRS + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DEEP_ENGAGEMENT + CHICAGO_HOME, data = .)
Residuals:
Min 1Q Median 3Q Max
-3.3104 -0.5563 0.1455 0.7466 3.8405
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.75738 0.02334 75.294 < 2e-16 ***
ACTIVE_PROPOSALS 0.24164 0.03801 6.356 2.11e-10 ***
AGE 0.29278 0.01779 16.459 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.24898 0.04006 6.215 5.24e-10 ***
VISITS_5PLUS 0.71995 0.02876 25.029 < 2e-16 ***
AF_25K_GIFT 0.88025 0.05003 17.595 < 2e-16 ***
GAVE_IN_LAST_3_YRS 0.77776 0.01937 40.146 < 2e-16 ***
MG_250K_PLUS 1.75820 0.08443 20.824 < 2e-16 ***
PRESIDENT_VISIT 0.15414 0.07058 2.184 0.029 *
TRUSTEE_OR_ADVISORY_BOARD 0.35971 0.03770 9.541 < 2e-16 ***
Alumnus 0.02544 0.02303 1.105 0.269
DEEP_ENGAGEMENT 0.18433 0.01908 9.660 < 2e-16 ***
CHICAGO_HOME 0.13271 0.01859 7.140 9.65e-13 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 1.116 on 18761 degrees of freedom
Multiple R-squared: 0.3313, Adjusted R-squared: 0.3308
F-statistic: 774.5 on 12 and 18761 DF, p-value: < 2.2e-16Transaction all predictors 1
lapply(tlmaps, function(x) summary(x))[[1]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ACTIVE_PROPOSALS +
ns(NUMERIC_AGE, df = 5) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DOUBLE_ALUM + EVER_PARENT + SEASON_TICKET_YEARS +
CHICAGO_HOME + QUAL_LEVEL + AFFINITY_SCORE + MG_PR_MODEL_DESC,
data = .)
Residuals:
Min 1Q Median 3Q Max
-2.7053 -0.6568 -0.0170 0.5699 4.3328
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.021955 0.101441 10.074 < 2e-16 ***
ACTIVE_PROPOSALS 0.064916 0.029946 2.168 0.030190 *
ns(NUMERIC_AGE, df = 5)1 0.184741 0.078192 2.363 0.018155 *
ns(NUMERIC_AGE, df = 5)2 0.020354 0.091488 0.222 0.823946
ns(NUMERIC_AGE, df = 5)3 -0.296885 0.062743 -4.732 2.24e-06 ***
ns(NUMERIC_AGE, df = 5)4 0.370226 0.193198 1.916 0.055341 .
ns(NUMERIC_AGE, df = 5)5 -0.066365 0.103325 -0.642 0.520690
PM_VISIT_LAST_2_YRS 0.126677 0.032031 3.955 7.69e-05 ***
log10plus1(VISIT_COUNT) 0.482156 0.027181 17.739 < 2e-16 ***
AF_25K_GIFT 0.631661 0.039377 16.041 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.440404 0.006149 71.620 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 0.413020 0.038425 10.749 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 -0.538985 0.023122 -23.311 < 2e-16 ***
MG_250K_PLUS 1.415328 0.070100 20.190 < 2e-16 ***
PRESIDENT_VISIT 0.054140 0.056502 0.958 0.337977
TRUSTEE_OR_ADVISORY_BOARD 0.092482 0.029565 3.128 0.001762 **
Alumnus -0.651132 0.021967 -29.642 < 2e-16 ***
DOUBLE_ALUM 0.026994 0.020840 1.295 0.195231
EVER_PARENT -0.079988 0.019923 -4.015 5.97e-05 ***
SEASON_TICKET_YEARS -0.019589 0.004086 -4.795 1.64e-06 ***
CHICAGO_HOME 0.052070 0.015205 3.425 0.000617 ***
QUAL_LEVELA1 $100M+ 1.215117 0.440200 2.760 0.005779 **
QUAL_LEVELA2 $50M - 99.9M 0.899936 0.393330 2.288 0.022149 *
QUAL_LEVELA3 $25M - $49.9M 0.058131 0.196288 0.296 0.767118
QUAL_LEVELA4 $10M - $24.9M 0.528339 0.135672 3.894 9.88e-05 ***
QUAL_LEVELA5 $5M - $9.9M 0.524364 0.102983 5.092 3.58e-07 ***
QUAL_LEVELA6 $2M - $4.9M 0.417957 0.093096 4.490 7.18e-06 ***
QUAL_LEVELA7 $1M - $1.9M 0.351987 0.073720 4.775 1.81e-06 ***
QUAL_LEVELB $500K - $999K 0.177756 0.066557 2.671 0.007575 **
QUAL_LEVELC $250K - $499K 0.093235 0.063005 1.480 0.138941
QUAL_LEVELD $100K - $249K 0.036398 0.062122 0.586 0.557939
QUAL_LEVELE $50K - $99K 0.076065 0.063542 1.197 0.231292
QUAL_LEVELF $25K - $49K 0.072746 0.063740 1.141 0.253763
QUAL_LEVELG $10K - $24K -0.027438 0.063047 -0.435 0.663428
QUAL_LEVELH Under $10K 0.011597 0.202922 0.057 0.954426
QUAL_LEVELJ Future Prospect -0.574098 0.616611 -0.931 0.351838
AFFINITY_SCORE 0.082169 0.006523 12.596 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier 0.106857 0.020745 5.151 2.62e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.107075 0.026045 4.111 3.95e-05 ***
MG_PR_MODEL_DESCTop Tier 0.325651 0.026368 12.350 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.867 on 18738 degrees of freedom
Multiple R-squared: 0.5967, Adjusted R-squared: 0.5958
F-statistic: 710.8 on 39 and 18738 DF, p-value: < 2.2e-16
[[2]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ACTIVE_PROPOSALS +
ns(NUMERIC_AGE, df = 5) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DOUBLE_ALUM + EVER_PARENT + SEASON_TICKET_YEARS +
CHICAGO_HOME + QUAL_LEVEL + AFFINITY_SCORE + MG_PR_MODEL_DESC,
data = .)
Residuals:
Min 1Q Median 3Q Max
-3.1745 -0.6505 -0.0165 0.5678 4.2902
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.066102 0.100013 10.660 < 2e-16 ***
ACTIVE_PROPOSALS 0.079937 0.030049 2.660 0.007815 **
ns(NUMERIC_AGE, df = 5)1 0.171675 0.077135 2.226 0.026050 *
ns(NUMERIC_AGE, df = 5)2 0.022380 0.090368 0.248 0.804406
ns(NUMERIC_AGE, df = 5)3 -0.321323 0.061160 -5.254 1.51e-07 ***
ns(NUMERIC_AGE, df = 5)4 0.371307 0.189527 1.959 0.050113 .
ns(NUMERIC_AGE, df = 5)5 -0.049503 0.093959 -0.527 0.598295
PM_VISIT_LAST_2_YRS 0.118800 0.032053 3.706 0.000211 ***
log10plus1(VISIT_COUNT) 0.455210 0.027168 16.756 < 2e-16 ***
AF_25K_GIFT 0.590294 0.039856 14.811 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.441899 0.006123 72.176 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 0.399268 0.038306 10.423 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 -0.557109 0.023096 -24.122 < 2e-16 ***
MG_250K_PLUS 1.406444 0.068493 20.534 < 2e-16 ***
PRESIDENT_VISIT 0.047047 0.056337 0.835 0.403673
TRUSTEE_OR_ADVISORY_BOARD 0.088780 0.029713 2.988 0.002812 **
Alumnus -0.649609 0.021897 -29.666 < 2e-16 ***
DOUBLE_ALUM 0.020791 0.020657 1.006 0.314192
EVER_PARENT -0.079549 0.019828 -4.012 6.04e-05 ***
SEASON_TICKET_YEARS -0.018450 0.004110 -4.489 7.19e-06 ***
CHICAGO_HOME 0.052021 0.015147 3.434 0.000595 ***
QUAL_LEVELA1 $100M+ 0.647735 0.392745 1.649 0.099113 .
QUAL_LEVELA2 $50M - 99.9M 0.561881 0.358652 1.567 0.117215
QUAL_LEVELA3 $25M - $49.9M 0.172196 0.196342 0.877 0.380486
QUAL_LEVELA4 $10M - $24.9M 0.683036 0.139451 4.898 9.76e-07 ***
QUAL_LEVELA5 $5M - $9.9M 0.452690 0.100601 4.500 6.84e-06 ***
QUAL_LEVELA6 $2M - $4.9M 0.390070 0.092543 4.215 2.51e-05 ***
QUAL_LEVELA7 $1M - $1.9M 0.314154 0.073337 4.284 1.85e-05 ***
QUAL_LEVELB $500K - $999K 0.114483 0.066024 1.734 0.082942 .
QUAL_LEVELC $250K - $499K 0.030799 0.062501 0.493 0.622175
QUAL_LEVELD $100K - $249K -0.023645 0.061588 -0.384 0.701042
QUAL_LEVELE $50K - $99K 0.019218 0.063035 0.305 0.760463
QUAL_LEVELF $25K - $49K 0.020928 0.063238 0.331 0.740695
QUAL_LEVELG $10K - $24K -0.075554 0.062495 -1.209 0.226699
QUAL_LEVELH Under $10K -0.144356 0.212210 -0.680 0.496356
QUAL_LEVELJ Future Prospect -0.637646 0.614744 -1.037 0.299630
AFFINITY_SCORE 0.085044 0.006504 13.076 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier 0.114493 0.020685 5.535 3.15e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.132334 0.025974 5.095 3.52e-07 ***
MG_PR_MODEL_DESCTop Tier 0.361791 0.026449 13.679 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8644 on 18738 degrees of freedom
Multiple R-squared: 0.6007, Adjusted R-squared: 0.5999
F-statistic: 722.9 on 39 and 18738 DF, p-value: < 2.2e-16
[[3]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ACTIVE_PROPOSALS +
ns(NUMERIC_AGE, df = 5) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DOUBLE_ALUM + EVER_PARENT + SEASON_TICKET_YEARS +
CHICAGO_HOME + QUAL_LEVEL + AFFINITY_SCORE + MG_PR_MODEL_DESC,
data = .)
Residuals:
Min 1Q Median 3Q Max
-3.1278 -0.6566 -0.0156 0.5675 4.2994
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.086437 0.102040 10.647 < 2e-16 ***
ACTIVE_PROPOSALS 0.064324 0.030158 2.133 0.032946 *
ns(NUMERIC_AGE, df = 5)1 0.160076 0.078909 2.029 0.042512 *
ns(NUMERIC_AGE, df = 5)2 -0.024792 0.092411 -0.268 0.788485
ns(NUMERIC_AGE, df = 5)3 -0.302639 0.062854 -4.815 1.48e-06 ***
ns(NUMERIC_AGE, df = 5)4 0.280442 0.195058 1.438 0.150525
ns(NUMERIC_AGE, df = 5)5 -0.090808 0.102840 -0.883 0.377247
PM_VISIT_LAST_2_YRS 0.131969 0.032261 4.091 4.32e-05 ***
log10plus1(VISIT_COUNT) 0.473675 0.027180 17.427 < 2e-16 ***
AF_25K_GIFT 0.619364 0.039519 15.672 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.443350 0.006152 72.068 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 0.391987 0.038436 10.198 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 -0.553741 0.023220 -23.848 < 2e-16 ***
MG_250K_PLUS 1.427278 0.069665 20.488 < 2e-16 ***
PRESIDENT_VISIT 0.041928 0.056175 0.746 0.455446
TRUSTEE_OR_ADVISORY_BOARD 0.096112 0.029844 3.220 0.001282 **
Alumnus -0.650304 0.021957 -29.618 < 2e-16 ***
DOUBLE_ALUM 0.026954 0.020762 1.298 0.194214
EVER_PARENT -0.068840 0.019945 -3.452 0.000559 ***
SEASON_TICKET_YEARS -0.019539 0.004046 -4.829 1.38e-06 ***
CHICAGO_HOME 0.053142 0.015264 3.481 0.000500 ***
QUAL_LEVELA1 $100M+ 0.643544 0.394718 1.630 0.103036
QUAL_LEVELA2 $50M - 99.9M 0.474391 0.439695 1.079 0.280642
QUAL_LEVELA3 $25M - $49.9M 0.185759 0.213986 0.868 0.385356
QUAL_LEVELA4 $10M - $24.9M 0.487373 0.136105 3.581 0.000343 ***
QUAL_LEVELA5 $5M - $9.9M 0.440485 0.101038 4.360 1.31e-05 ***
QUAL_LEVELA6 $2M - $4.9M 0.326500 0.092158 3.543 0.000397 ***
QUAL_LEVELA7 $1M - $1.9M 0.314727 0.074140 4.245 2.20e-05 ***
QUAL_LEVELB $500K - $999K 0.111854 0.066415 1.684 0.092164 .
QUAL_LEVELC $250K - $499K 0.043985 0.062890 0.699 0.484313
QUAL_LEVELD $100K - $249K -0.006485 0.061981 -0.105 0.916666
QUAL_LEVELE $50K - $99K 0.041382 0.063457 0.652 0.514324
QUAL_LEVELF $25K - $49K 0.019167 0.063613 0.301 0.763181
QUAL_LEVELG $10K - $24K -0.082945 0.062915 -1.318 0.187393
QUAL_LEVELH Under $10K -0.046328 0.203245 -0.228 0.819695
QUAL_LEVELJ Future Prospect -0.628660 0.617795 -1.018 0.308887
AFFINITY_SCORE 0.086452 0.006532 13.235 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier 0.111683 0.020772 5.377 7.68e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.109709 0.026130 4.199 2.70e-05 ***
MG_PR_MODEL_DESCTop Tier 0.333230 0.026504 12.573 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8687 on 18738 degrees of freedom
Multiple R-squared: 0.599, Adjusted R-squared: 0.5982
F-statistic: 717.8 on 39 and 18738 DF, p-value: < 2.2e-16
[[4]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ACTIVE_PROPOSALS +
ns(NUMERIC_AGE, df = 5) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DOUBLE_ALUM + EVER_PARENT + SEASON_TICKET_YEARS +
CHICAGO_HOME + QUAL_LEVEL + AFFINITY_SCORE + MG_PR_MODEL_DESC,
data = .)
Residuals:
Min 1Q Median 3Q Max
-3.1385 -0.6477 -0.0173 0.5656 4.2500
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.075776 0.101189 10.631 < 2e-16 ***
ACTIVE_PROPOSALS 0.056357 0.030121 1.871 0.061359 .
ns(NUMERIC_AGE, df = 5)1 0.208437 0.077625 2.685 0.007256 **
ns(NUMERIC_AGE, df = 5)2 0.046877 0.090814 0.516 0.605731
ns(NUMERIC_AGE, df = 5)3 -0.271445 0.062506 -4.343 1.41e-05 ***
ns(NUMERIC_AGE, df = 5)4 0.436106 0.191861 2.273 0.023036 *
ns(NUMERIC_AGE, df = 5)5 -0.091444 0.103066 -0.887 0.374960
PM_VISIT_LAST_2_YRS 0.136709 0.032271 4.236 2.28e-05 ***
log10plus1(VISIT_COUNT) 0.466296 0.027101 17.206 < 2e-16 ***
AF_25K_GIFT 0.638429 0.039410 16.200 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.437477 0.006140 71.250 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 0.417345 0.038257 10.909 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 -0.542582 0.023043 -23.547 < 2e-16 ***
MG_250K_PLUS 1.367698 0.069519 19.674 < 2e-16 ***
PRESIDENT_VISIT 0.088922 0.056649 1.570 0.116503
TRUSTEE_OR_ADVISORY_BOARD 0.106150 0.029852 3.556 0.000378 ***
Alumnus -0.662163 0.021901 -30.234 < 2e-16 ***
DOUBLE_ALUM 0.023889 0.020738 1.152 0.249369
EVER_PARENT -0.076855 0.019788 -3.884 0.000103 ***
SEASON_TICKET_YEARS -0.020464 0.004049 -5.054 4.37e-07 ***
CHICAGO_HOME 0.054466 0.015196 3.584 0.000339 ***
QUAL_LEVELA1 $100M+ 0.598298 0.393617 1.520 0.128528
QUAL_LEVELA2 $50M - 99.9M 0.545790 0.359455 1.518 0.128935
QUAL_LEVELA3 $25M - $49.9M 0.051140 0.196861 0.260 0.795039
QUAL_LEVELA4 $10M - $24.9M 0.518233 0.133633 3.878 0.000106 ***
QUAL_LEVELA5 $5M - $9.9M 0.447974 0.100738 4.447 8.76e-06 ***
QUAL_LEVELA6 $2M - $4.9M 0.312072 0.094260 3.311 0.000932 ***
QUAL_LEVELA7 $1M - $1.9M 0.283104 0.073930 3.829 0.000129 ***
QUAL_LEVELB $500K - $999K 0.114047 0.066556 1.714 0.086627 .
QUAL_LEVELC $250K - $499K 0.022118 0.062963 0.351 0.725384
QUAL_LEVELD $100K - $249K -0.026800 0.062075 -0.432 0.665934
QUAL_LEVELE $50K - $99K 0.011933 0.063540 0.188 0.851037
QUAL_LEVELF $25K - $49K 0.014054 0.063731 0.221 0.825464
QUAL_LEVELG $10K - $24K -0.089411 0.063059 -1.418 0.156239
QUAL_LEVELH Under $10K -0.094743 0.207598 -0.456 0.648122
QUAL_LEVELJ Future Prospect -0.539164 0.868765 -0.621 0.534864
AFFINITY_SCORE 0.083320 0.006522 12.775 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier 0.095045 0.020754 4.580 4.69e-06 ***
MG_PR_MODEL_DESCMiddle Tier 0.100137 0.026102 3.836 0.000125 ***
MG_PR_MODEL_DESCTop Tier 0.334006 0.026433 12.636 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8661 on 18738 degrees of freedom
Multiple R-squared: 0.5966, Adjusted R-squared: 0.5957
F-statistic: 710.5 on 39 and 18738 DF, p-value: < 2.2e-16
[[5]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ACTIVE_PROPOSALS +
ns(NUMERIC_AGE, df = 5) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DOUBLE_ALUM + EVER_PARENT + SEASON_TICKET_YEARS +
CHICAGO_HOME + QUAL_LEVEL + AFFINITY_SCORE + MG_PR_MODEL_DESC,
data = .)
Residuals:
Min 1Q Median 3Q Max
-3.1457 -0.6529 -0.0178 0.5671 4.3115
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.017443 0.101075 10.066 < 2e-16 ***
ACTIVE_PROPOSALS 0.058092 0.030271 1.919 0.054987 .
ns(NUMERIC_AGE, df = 5)1 0.195604 0.078173 2.502 0.012351 *
ns(NUMERIC_AGE, df = 5)2 0.041959 0.091551 0.458 0.646734
ns(NUMERIC_AGE, df = 5)3 -0.317950 0.062817 -5.062 4.20e-07 ***
ns(NUMERIC_AGE, df = 5)4 0.404442 0.193538 2.090 0.036656 *
ns(NUMERIC_AGE, df = 5)5 -0.083635 0.104361 -0.801 0.422907
PM_VISIT_LAST_2_YRS 0.143388 0.032307 4.438 9.12e-06 ***
log10plus1(VISIT_COUNT) 0.469516 0.027281 17.210 < 2e-16 ***
AF_25K_GIFT 0.635891 0.039555 16.076 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.443313 0.006132 72.298 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 0.402426 0.038376 10.486 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 -0.549712 0.023167 -23.728 < 2e-16 ***
MG_250K_PLUS 1.380573 0.069561 19.847 < 2e-16 ***
PRESIDENT_VISIT 0.058229 0.056022 1.039 0.298633
TRUSTEE_OR_ADVISORY_BOARD 0.094628 0.029678 3.189 0.001432 **
Alumnus -0.658483 0.021956 -29.991 < 2e-16 ***
DOUBLE_ALUM 0.030263 0.020917 1.447 0.147966
EVER_PARENT -0.076887 0.019963 -3.851 0.000118 ***
SEASON_TICKET_YEARS -0.020476 0.004106 -4.987 6.18e-07 ***
CHICAGO_HOME 0.056806 0.015195 3.738 0.000186 ***
QUAL_LEVELA1 $100M+ 0.660152 0.394234 1.675 0.094046 .
QUAL_LEVELA2 $50M - 99.9M 0.589679 0.360033 1.638 0.101471
QUAL_LEVELA3 $25M - $49.9M 0.168707 0.209285 0.806 0.420189
QUAL_LEVELA4 $10M - $24.9M 0.524055 0.134585 3.894 9.90e-05 ***
QUAL_LEVELA5 $5M - $9.9M 0.493427 0.101009 4.885 1.04e-06 ***
QUAL_LEVELA6 $2M - $4.9M 0.351755 0.091754 3.834 0.000127 ***
QUAL_LEVELA7 $1M - $1.9M 0.347161 0.074195 4.679 2.90e-06 ***
QUAL_LEVELB $500K - $999K 0.150040 0.066751 2.248 0.024603 *
QUAL_LEVELC $250K - $499K 0.062887 0.063226 0.995 0.319924
QUAL_LEVELD $100K - $249K 0.007581 0.062327 0.122 0.903196
QUAL_LEVELE $50K - $99K 0.045734 0.063757 0.717 0.473190
QUAL_LEVELF $25K - $49K 0.041980 0.064004 0.656 0.511902
QUAL_LEVELG $10K - $24K -0.061653 0.063198 -0.976 0.329295
QUAL_LEVELH Under $10K -0.064750 0.208030 -0.311 0.755612
QUAL_LEVELJ Future Prospect -0.616339 0.617007 -0.999 0.317848
AFFINITY_SCORE 0.086197 0.006514 13.233 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier 0.119110 0.020762 5.737 9.80e-09 ***
MG_PR_MODEL_DESCMiddle Tier 0.110832 0.026037 4.257 2.08e-05 ***
MG_PR_MODEL_DESCTop Tier 0.334177 0.026393 12.662 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8675 on 18738 degrees of freedom
Multiple R-squared: 0.5976, Adjusted R-squared: 0.5968
F-statistic: 713.5 on 39 and 18738 DF, p-value: < 2.2e-16
[[6]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ACTIVE_PROPOSALS +
ns(NUMERIC_AGE, df = 5) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DOUBLE_ALUM + EVER_PARENT + SEASON_TICKET_YEARS +
CHICAGO_HOME + QUAL_LEVEL + AFFINITY_SCORE + MG_PR_MODEL_DESC,
data = .)
Residuals:
Min 1Q Median 3Q Max
-3.1443 -0.6546 -0.0173 0.5677 4.3298
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.014074 0.101487 9.992 < 2e-16 ***
ACTIVE_PROPOSALS 0.077205 0.030203 2.556 0.010590 *
ns(NUMERIC_AGE, df = 5)1 0.199207 0.077457 2.572 0.010124 *
ns(NUMERIC_AGE, df = 5)2 0.018731 0.090691 0.207 0.836377
ns(NUMERIC_AGE, df = 5)3 -0.298376 0.062630 -4.764 1.91e-06 ***
ns(NUMERIC_AGE, df = 5)4 0.370236 0.191737 1.931 0.053503 .
ns(NUMERIC_AGE, df = 5)5 -0.081337 0.103847 -0.783 0.433494
PM_VISIT_LAST_2_YRS 0.126247 0.032356 3.902 9.58e-05 ***
log10plus1(VISIT_COUNT) 0.471299 0.027197 17.329 < 2e-16 ***
AF_25K_GIFT 0.621000 0.040161 15.463 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.444489 0.006157 72.187 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 0.410265 0.038479 10.662 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 -0.545219 0.023137 -23.565 < 2e-16 ***
MG_250K_PLUS 1.429444 0.070625 20.240 < 2e-16 ***
PRESIDENT_VISIT 0.092413 0.057064 1.619 0.105368
TRUSTEE_OR_ADVISORY_BOARD 0.092666 0.029721 3.118 0.001824 **
Alumnus -0.663096 0.021946 -30.215 < 2e-16 ***
DOUBLE_ALUM 0.021844 0.020822 1.049 0.294142
EVER_PARENT -0.079606 0.019927 -3.995 6.49e-05 ***
SEASON_TICKET_YEARS -0.020793 0.004066 -5.114 3.18e-07 ***
CHICAGO_HOME 0.056936 0.015197 3.746 0.000180 ***
QUAL_LEVELA1 $100M+ 0.558186 0.439163 1.271 0.203737
QUAL_LEVELA2 $50M - 99.9M 0.607994 0.360193 1.688 0.091434 .
QUAL_LEVELA3 $25M - $49.9M 0.118317 0.193583 0.611 0.541079
QUAL_LEVELA4 $10M - $24.9M 0.531333 0.137850 3.854 0.000116 ***
QUAL_LEVELA5 $5M - $9.9M 0.548998 0.102961 5.332 9.82e-08 ***
QUAL_LEVELA6 $2M - $4.9M 0.367464 0.094332 3.895 9.84e-05 ***
QUAL_LEVELA7 $1M - $1.9M 0.361506 0.074969 4.822 1.43e-06 ***
QUAL_LEVELB $500K - $999K 0.169760 0.067755 2.505 0.012237 *
QUAL_LEVELC $250K - $499K 0.096903 0.064182 1.510 0.131111
QUAL_LEVELD $100K - $249K 0.049975 0.063331 0.789 0.430055
QUAL_LEVELE $50K - $99K 0.075665 0.064760 1.168 0.242663
QUAL_LEVELF $25K - $49K 0.082905 0.064965 1.276 0.201921
QUAL_LEVELG $10K - $24K -0.004286 0.064242 -0.067 0.946812
QUAL_LEVELH Under $10K -0.023687 0.213432 -0.111 0.911634
QUAL_LEVELJ Future Prospect -0.564666 0.617079 -0.915 0.360170
AFFINITY_SCORE 0.082990 0.006537 12.696 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier 0.098747 0.020811 4.745 2.10e-06 ***
MG_PR_MODEL_DESCMiddle Tier 0.098389 0.026119 3.767 0.000166 ***
MG_PR_MODEL_DESCTop Tier 0.321191 0.026428 12.153 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8675 on 18738 degrees of freedom
Multiple R-squared: 0.5972, Adjusted R-squared: 0.5963
F-statistic: 712.2 on 39 and 18738 DF, p-value: < 2.2e-16
[[7]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ACTIVE_PROPOSALS +
ns(NUMERIC_AGE, df = 5) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DOUBLE_ALUM + EVER_PARENT + SEASON_TICKET_YEARS +
CHICAGO_HOME + QUAL_LEVEL + AFFINITY_SCORE + MG_PR_MODEL_DESC,
data = .)
Residuals:
Min 1Q Median 3Q Max
-3.0940 -0.6544 -0.0166 0.5659 4.2983
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.068654 0.100047 10.681 < 2e-16 ***
ACTIVE_PROPOSALS 0.041140 0.030204 1.362 0.173194
ns(NUMERIC_AGE, df = 5)1 0.163003 0.077090 2.114 0.034491 *
ns(NUMERIC_AGE, df = 5)2 0.002642 0.090354 0.029 0.976669
ns(NUMERIC_AGE, df = 5)3 -0.335122 0.062730 -5.342 9.29e-08 ***
ns(NUMERIC_AGE, df = 5)4 0.333280 0.191010 1.745 0.081030 .
ns(NUMERIC_AGE, df = 5)5 -0.088111 0.105052 -0.839 0.401627
PM_VISIT_LAST_2_YRS 0.132220 0.032201 4.106 4.04e-05 ***
log10plus1(VISIT_COUNT) 0.477731 0.027194 17.567 < 2e-16 ***
AF_25K_GIFT 0.637426 0.039471 16.149 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.444150 0.006142 72.308 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 0.401218 0.038443 10.437 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 -0.541001 0.023160 -23.359 < 2e-16 ***
MG_250K_PLUS 1.439520 0.070275 20.484 < 2e-16 ***
PRESIDENT_VISIT 0.098560 0.056371 1.748 0.080407 .
TRUSTEE_OR_ADVISORY_BOARD 0.098376 0.029591 3.324 0.000887 ***
Alumnus -0.655582 0.022033 -29.755 < 2e-16 ***
DOUBLE_ALUM 0.025264 0.020952 1.206 0.227898
EVER_PARENT -0.084963 0.020016 -4.245 2.20e-05 ***
SEASON_TICKET_YEARS -0.020139 0.004141 -4.864 1.16e-06 ***
CHICAGO_HOME 0.058188 0.015197 3.829 0.000129 ***
QUAL_LEVELA1 $100M+ 0.625953 0.394281 1.588 0.112398
QUAL_LEVELA2 $50M - 99.9M 0.338446 0.438767 0.771 0.440505
QUAL_LEVELA3 $25M - $49.9M 0.086432 0.204667 0.422 0.672808
QUAL_LEVELA4 $10M - $24.9M 0.613404 0.136053 4.509 6.57e-06 ***
QUAL_LEVELA5 $5M - $9.9M 0.566557 0.100455 5.640 1.73e-08 ***
QUAL_LEVELA6 $2M - $4.9M 0.359408 0.093190 3.857 0.000115 ***
QUAL_LEVELA7 $1M - $1.9M 0.303852 0.073585 4.129 3.66e-05 ***
QUAL_LEVELB $500K - $999K 0.137050 0.066478 2.062 0.039262 *
QUAL_LEVELC $250K - $499K 0.040092 0.062854 0.638 0.523579
QUAL_LEVELD $100K - $249K 0.001632 0.061968 0.026 0.978983
QUAL_LEVELE $50K - $99K 0.033097 0.063415 0.522 0.601740
QUAL_LEVELF $25K - $49K 0.022298 0.063656 0.350 0.726124
QUAL_LEVELG $10K - $24K -0.072604 0.062864 -1.155 0.248133
QUAL_LEVELH Under $10K 0.079695 0.218783 0.364 0.715663
QUAL_LEVELJ Future Prospect -0.702462 0.870678 -0.807 0.419793
AFFINITY_SCORE 0.083285 0.006525 12.764 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier 0.116635 0.020786 5.611 2.04e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.116749 0.026127 4.468 7.92e-06 ***
MG_PR_MODEL_DESCTop Tier 0.330310 0.026447 12.489 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8677 on 18738 degrees of freedom
Multiple R-squared: 0.5971, Adjusted R-squared: 0.5962
F-statistic: 711.9 on 39 and 18738 DF, p-value: < 2.2e-16
[[8]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ACTIVE_PROPOSALS +
ns(NUMERIC_AGE, df = 5) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DOUBLE_ALUM + EVER_PARENT + SEASON_TICKET_YEARS +
CHICAGO_HOME + QUAL_LEVEL + AFFINITY_SCORE + MG_PR_MODEL_DESC,
data = .)
Residuals:
Min 1Q Median 3Q Max
-3.1610 -0.6578 -0.0152 0.5677 4.2431
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.054415 0.101893 10.348 < 2e-16 ***
ACTIVE_PROPOSALS 0.061617 0.030132 2.045 0.040879 *
ns(NUMERIC_AGE, df = 5)1 0.222418 0.078000 2.852 0.004356 **
ns(NUMERIC_AGE, df = 5)2 0.086043 0.091353 0.942 0.346273
ns(NUMERIC_AGE, df = 5)3 -0.284517 0.062578 -4.547 5.49e-06 ***
ns(NUMERIC_AGE, df = 5)4 0.495247 0.192828 2.568 0.010226 *
ns(NUMERIC_AGE, df = 5)5 -0.009570 0.102481 -0.093 0.925598
PM_VISIT_LAST_2_YRS 0.140860 0.032363 4.353 1.35e-05 ***
log10plus1(VISIT_COUNT) 0.477234 0.027253 17.511 < 2e-16 ***
AF_25K_GIFT 0.598064 0.039938 14.975 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.439166 0.006163 71.262 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 0.404708 0.038292 10.569 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 -0.543131 0.023187 -23.424 < 2e-16 ***
MG_250K_PLUS 1.379719 0.069845 19.754 < 2e-16 ***
PRESIDENT_VISIT 0.063967 0.057157 1.119 0.263091
TRUSTEE_OR_ADVISORY_BOARD 0.088990 0.030052 2.961 0.003068 **
Alumnus -0.649191 0.022011 -29.494 < 2e-16 ***
DOUBLE_ALUM 0.033747 0.020765 1.625 0.104135
EVER_PARENT -0.087326 0.019958 -4.376 1.22e-05 ***
SEASON_TICKET_YEARS -0.017432 0.004079 -4.273 1.93e-05 ***
CHICAGO_HOME 0.052681 0.015232 3.459 0.000544 ***
QUAL_LEVELA1 $100M+ 0.144561 0.439016 0.329 0.741945
QUAL_LEVELA2 $50M - 99.9M 0.527219 0.360149 1.464 0.143241
QUAL_LEVELA3 $25M - $49.9M 0.041923 0.213887 0.196 0.844609
QUAL_LEVELA4 $10M - $24.9M 0.501768 0.135263 3.710 0.000208 ***
QUAL_LEVELA5 $5M - $9.9M 0.521536 0.104312 5.000 5.79e-07 ***
QUAL_LEVELA6 $2M - $4.9M 0.306747 0.093858 3.268 0.001084 **
QUAL_LEVELA7 $1M - $1.9M 0.284876 0.075140 3.791 0.000150 ***
QUAL_LEVELB $500K - $999K 0.080796 0.067740 1.193 0.232991
QUAL_LEVELC $250K - $499K 0.009287 0.064188 0.145 0.884960
QUAL_LEVELD $100K - $249K -0.051857 0.063318 -0.819 0.412800
QUAL_LEVELE $50K - $99K -0.021342 0.064749 -0.330 0.741705
QUAL_LEVELF $25K - $49K -0.020085 0.064950 -0.309 0.757144
QUAL_LEVELG $10K - $24K -0.103553 0.064209 -1.613 0.106813
QUAL_LEVELH Under $10K -0.072198 0.219189 -0.329 0.741865
QUAL_LEVELJ Future Prospect -0.667151 0.616985 -1.081 0.279574
AFFINITY_SCORE 0.084861 0.006535 12.986 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier 0.103630 0.020718 5.002 5.73e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.107867 0.026056 4.140 3.49e-05 ***
MG_PR_MODEL_DESCTop Tier 0.330270 0.026424 12.499 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8673 on 18738 degrees of freedom
Multiple R-squared: 0.5966, Adjusted R-squared: 0.5957
F-statistic: 710.5 on 39 and 18738 DF, p-value: < 2.2e-16
[[9]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ACTIVE_PROPOSALS +
ns(NUMERIC_AGE, df = 5) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DOUBLE_ALUM + EVER_PARENT + SEASON_TICKET_YEARS +
CHICAGO_HOME + QUAL_LEVEL + AFFINITY_SCORE + MG_PR_MODEL_DESC,
data = .)
Residuals:
Min 1Q Median 3Q Max
-3.1443 -0.6507 -0.0139 0.5688 4.2863
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.018995 0.100703 10.119 < 2e-16 ***
ACTIVE_PROPOSALS 0.064438 0.029814 2.161 0.030682 *
ns(NUMERIC_AGE, df = 5)1 0.221519 0.077625 2.854 0.004326 **
ns(NUMERIC_AGE, df = 5)2 0.077295 0.090935 0.850 0.395338
ns(NUMERIC_AGE, df = 5)3 -0.283940 0.062605 -4.535 5.78e-06 ***
ns(NUMERIC_AGE, df = 5)4 0.452125 0.191914 2.356 0.018489 *
ns(NUMERIC_AGE, df = 5)5 -0.081511 0.103765 -0.786 0.432153
PM_VISIT_LAST_2_YRS 0.126050 0.031899 3.952 7.79e-05 ***
log10plus1(VISIT_COUNT) 0.462359 0.027172 17.016 < 2e-16 ***
AF_25K_GIFT 0.615762 0.039599 15.550 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.444080 0.006154 72.159 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 0.403265 0.038539 10.464 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 -0.535904 0.023131 -23.169 < 2e-16 ***
MG_250K_PLUS 1.372640 0.069487 19.754 < 2e-16 ***
PRESIDENT_VISIT 0.052422 0.056079 0.935 0.349905
TRUSTEE_OR_ADVISORY_BOARD 0.089180 0.029712 3.001 0.002690 **
Alumnus -0.654223 0.021953 -29.802 < 2e-16 ***
DOUBLE_ALUM 0.020569 0.020818 0.988 0.323142
EVER_PARENT -0.078573 0.019965 -3.936 8.33e-05 ***
SEASON_TICKET_YEARS -0.021763 0.004069 -5.348 8.98e-08 ***
CHICAGO_HOME 0.053221 0.015200 3.501 0.000464 ***
QUAL_LEVELA1 $100M+ 0.706632 0.506194 1.396 0.162740
QUAL_LEVELA2 $50M - 99.9M 0.588582 0.359949 1.635 0.102028
QUAL_LEVELA3 $25M - $49.9M 0.045706 0.199717 0.229 0.818984
QUAL_LEVELA4 $10M - $24.9M 0.490708 0.134087 3.660 0.000253 ***
QUAL_LEVELA5 $5M - $9.9M 0.528860 0.101083 5.232 1.70e-07 ***
QUAL_LEVELA6 $2M - $4.9M 0.399074 0.092886 4.296 1.74e-05 ***
QUAL_LEVELA7 $1M - $1.9M 0.317383 0.074190 4.278 1.90e-05 ***
QUAL_LEVELB $500K - $999K 0.146891 0.066674 2.203 0.027599 *
QUAL_LEVELC $250K - $499K 0.047633 0.063239 0.753 0.451321
QUAL_LEVELD $100K - $249K -0.001135 0.062293 -0.018 0.985458
QUAL_LEVELE $50K - $99K 0.038863 0.063738 0.610 0.542043
QUAL_LEVELF $25K - $49K 0.040087 0.063968 0.627 0.530880
QUAL_LEVELG $10K - $24K -0.061519 0.063213 -0.973 0.330471
QUAL_LEVELH Under $10K -0.044074 0.198604 -0.222 0.824379
QUAL_LEVELJ Future Prospect -0.599734 0.616840 -0.972 0.330929
AFFINITY_SCORE 0.082056 0.006516 12.592 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier 0.094498 0.020826 4.537 5.73e-06 ***
MG_PR_MODEL_DESCMiddle Tier 0.095288 0.026207 3.636 0.000278 ***
MG_PR_MODEL_DESCTop Tier 0.336462 0.026394 12.747 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8673 on 18738 degrees of freedom
Multiple R-squared: 0.5984, Adjusted R-squared: 0.5975
F-statistic: 715.8 on 39 and 18738 DF, p-value: < 2.2e-16
[[10]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ACTIVE_PROPOSALS +
ns(NUMERIC_AGE, df = 5) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DOUBLE_ALUM + EVER_PARENT + SEASON_TICKET_YEARS +
CHICAGO_HOME + QUAL_LEVEL + AFFINITY_SCORE + MG_PR_MODEL_DESC,
data = .)
Residuals:
Min 1Q Median 3Q Max
-3.1430 -0.6636 -0.0128 0.5684 4.0394
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.984996 0.100246 9.826 < 2e-16 ***
ACTIVE_PROPOSALS 0.077020 0.030107 2.558 0.010530 *
ns(NUMERIC_AGE, df = 5)1 0.218233 0.077226 2.826 0.004720 **
ns(NUMERIC_AGE, df = 5)2 0.088870 0.090423 0.983 0.325703
ns(NUMERIC_AGE, df = 5)3 -0.294308 0.062442 -4.713 2.46e-06 ***
ns(NUMERIC_AGE, df = 5)4 0.507612 0.191032 2.657 0.007886 **
ns(NUMERIC_AGE, df = 5)5 -0.032575 0.102771 -0.317 0.751269
PM_VISIT_LAST_2_YRS 0.123971 0.032109 3.861 0.000113 ***
log10plus1(VISIT_COUNT) 0.464229 0.027202 17.066 < 2e-16 ***
AF_25K_GIFT 0.627301 0.039498 15.882 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.437671 0.006153 71.131 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 0.421697 0.038464 10.964 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 -0.525039 0.023153 -22.677 < 2e-16 ***
MG_250K_PLUS 1.397009 0.069481 20.106 < 2e-16 ***
PRESIDENT_VISIT 0.042544 0.056856 0.748 0.454300
TRUSTEE_OR_ADVISORY_BOARD 0.100446 0.029774 3.374 0.000743 ***
Alumnus -0.640247 0.021986 -29.121 < 2e-16 ***
DOUBLE_ALUM 0.012397 0.020834 0.595 0.551824
EVER_PARENT -0.077207 0.019983 -3.864 0.000112 ***
SEASON_TICKET_YEARS -0.020340 0.004114 -4.945 7.69e-07 ***
CHICAGO_HOME 0.057580 0.015249 3.776 0.000160 ***
QUAL_LEVELA1 $100M+ 0.659487 0.394690 1.671 0.094759 .
QUAL_LEVELA2 $50M - 99.9M 0.545122 0.393481 1.385 0.165952
QUAL_LEVELA3 $25M - $49.9M 0.171221 0.199877 0.857 0.391659
QUAL_LEVELA4 $10M - $24.9M 0.602859 0.134799 4.472 7.78e-06 ***
QUAL_LEVELA5 $5M - $9.9M 0.512281 0.101770 5.034 4.86e-07 ***
QUAL_LEVELA6 $2M - $4.9M 0.348640 0.091927 3.793 0.000150 ***
QUAL_LEVELA7 $1M - $1.9M 0.328067 0.074095 4.428 9.58e-06 ***
QUAL_LEVELB $500K - $999K 0.136262 0.066492 2.049 0.040449 *
QUAL_LEVELC $250K - $499K 0.068478 0.063056 1.086 0.277494
QUAL_LEVELD $100K - $249K 0.002294 0.062094 0.037 0.970528
QUAL_LEVELE $50K - $99K 0.046585 0.063577 0.733 0.463736
QUAL_LEVELF $25K - $49K 0.034613 0.063777 0.543 0.587324
QUAL_LEVELG $10K - $24K -0.054509 0.063017 -0.865 0.387049
QUAL_LEVELH Under $10K -0.041999 0.203290 -0.207 0.836328
QUAL_LEVELJ Future Prospect -0.597773 0.617731 -0.968 0.333211
AFFINITY_SCORE 0.084097 0.006525 12.888 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier 0.111407 0.020881 5.335 9.65e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.103621 0.026196 3.956 7.66e-05 ***
MG_PR_MODEL_DESCTop Tier 0.331219 0.026478 12.509 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8686 on 18734 degrees of freedom
Multiple R-squared: 0.5955, Adjusted R-squared: 0.5946
F-statistic: 707.1 on 39 and 18734 DF, p-value: < 2.2e-16## Transaction all predictors 2 {#appendix-trans-all-2}
lapply(tlmaps2, function(x) summary(x))[[1]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ns(NUMERIC_AGE,
df = 3) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + EVER_PARENT + SEASON_TICKET_YEARS + CHICAGO_HOME +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-2.8750 -0.6584 -0.0198 0.5747 4.2474
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.106745 0.066358 16.678 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)1 -0.453626 0.041248 -10.998 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)2 0.358644 0.140827 2.547 0.010883 *
ns(NUMERIC_AGE, df = 3)3 0.026666 0.079578 0.335 0.737563
PM_VISIT_LAST_2_YRS 0.192421 0.026059 7.384 1.60e-13 ***
log10plus1(VISIT_COUNT) 0.474494 0.026865 17.662 < 2e-16 ***
AF_25K_GIFT 0.684029 0.039212 17.444 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.438473 0.006157 71.214 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 0.422669 0.038528 10.970 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 -0.541653 0.023189 -23.358 < 2e-16 ***
MG_250K_PLUS 1.602063 0.066422 24.119 < 2e-16 ***
PRESIDENT_VISIT 0.157782 0.055195 2.859 0.004259 **
TRUSTEE_OR_ADVISORY_BOARD 0.102029 0.029641 3.442 0.000578 ***
Alumnus -0.644498 0.021767 -29.609 < 2e-16 ***
EVER_PARENT -0.076859 0.019829 -3.876 0.000106 ***
SEASON_TICKET_YEARS -0.019776 0.004100 -4.823 1.42e-06 ***
CHICAGO_HOME 0.051622 0.015248 3.386 0.000712 ***
AFFINITY_SCORE 0.077855 0.006422 12.124 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier 0.098617 0.020053 4.918 8.83e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.097681 0.025684 3.803 0.000143 ***
MG_PR_MODEL_DESCTop Tier 0.370779 0.025548 14.513 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8707 on 18757 degrees of freedom
Multiple R-squared: 0.5928, Adjusted R-squared: 0.5924
F-statistic: 1365 on 20 and 18757 DF, p-value: < 2.2e-16
[[2]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ns(NUMERIC_AGE,
df = 3) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + EVER_PARENT + SEASON_TICKET_YEARS + CHICAGO_HOME +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-2.8821 -0.6584 -0.0213 0.5727 4.2596
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.088929 0.066440 16.390 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)1 -0.441453 0.040704 -10.845 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)2 0.375457 0.139812 2.685 0.007250 **
ns(NUMERIC_AGE, df = 3)3 -0.004023 0.073173 -0.055 0.956151
PM_VISIT_LAST_2_YRS 0.191826 0.026036 7.368 1.81e-13 ***
log10plus1(VISIT_COUNT) 0.452768 0.026840 16.869 < 2e-16 ***
AF_25K_GIFT 0.645790 0.039735 16.252 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.439840 0.006126 71.797 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 0.407928 0.038410 10.620 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 -0.558673 0.023165 -24.117 < 2e-16 ***
MG_250K_PLUS 1.617986 0.064628 25.035 < 2e-16 ***
PRESIDENT_VISIT 0.164765 0.054930 3.000 0.002708 **
TRUSTEE_OR_ADVISORY_BOARD 0.099861 0.029783 3.353 0.000801 ***
Alumnus -0.648234 0.021709 -29.861 < 2e-16 ***
EVER_PARENT -0.076365 0.019731 -3.870 0.000109 ***
SEASON_TICKET_YEARS -0.018710 0.004126 -4.535 5.79e-06 ***
CHICAGO_HOME 0.050880 0.015188 3.350 0.000810 ***
AFFINITY_SCORE 0.080413 0.006400 12.565 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier 0.102848 0.019979 5.148 2.66e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.119968 0.025612 4.684 2.83e-06 ***
MG_PR_MODEL_DESCTop Tier 0.402678 0.025612 15.722 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8683 on 18757 degrees of freedom
Multiple R-squared: 0.5968, Adjusted R-squared: 0.5963
F-statistic: 1388 on 20 and 18757 DF, p-value: < 2.2e-16
[[3]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ns(NUMERIC_AGE,
df = 3) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + EVER_PARENT + SEASON_TICKET_YEARS + CHICAGO_HOME +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-2.8771 -0.6623 -0.0163 0.5748 4.2590
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.082645 0.067339 16.078 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)1 -0.449336 0.041525 -10.821 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)2 0.377538 0.142841 2.643 0.008223 **
ns(NUMERIC_AGE, df = 3)3 0.015328 0.078600 0.195 0.845382
PM_VISIT_LAST_2_YRS 0.197041 0.026073 7.557 4.30e-14 ***
log10plus1(VISIT_COUNT) 0.464230 0.026836 17.299 < 2e-16 ***
AF_25K_GIFT 0.670878 0.039352 17.048 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.441327 0.006155 71.707 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 0.399131 0.038524 10.361 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 -0.555852 0.023280 -23.877 < 2e-16 ***
MG_250K_PLUS 1.611500 0.065798 24.492 < 2e-16 ***
PRESIDENT_VISIT 0.141530 0.054885 2.579 0.009926 **
TRUSTEE_OR_ADVISORY_BOARD 0.106714 0.029899 3.569 0.000359 ***
Alumnus -0.642949 0.021747 -29.565 < 2e-16 ***
EVER_PARENT -0.065420 0.019842 -3.297 0.000979 ***
SEASON_TICKET_YEARS -0.019819 0.004058 -4.884 1.05e-06 ***
CHICAGO_HOME 0.053847 0.015299 3.520 0.000433 ***
AFFINITY_SCORE 0.081961 0.006425 12.757 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier 0.098145 0.020068 4.891 1.01e-06 ***
MG_PR_MODEL_DESCMiddle Tier 0.097822 0.025746 3.800 0.000145 ***
MG_PR_MODEL_DESCTop Tier 0.378193 0.025645 14.747 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.872 on 18757 degrees of freedom
Multiple R-squared: 0.5955, Adjusted R-squared: 0.5951
F-statistic: 1381 on 20 and 18757 DF, p-value: < 2.2e-16
[[4]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ns(NUMERIC_AGE,
df = 3) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + EVER_PARENT + SEASON_TICKET_YEARS + CHICAGO_HOME +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-2.8916 -0.6551 -0.0172 0.5712 4.2223
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.131272 0.065896 17.168 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)1 -0.460029 0.041068 -11.202 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)2 0.357113 0.139953 2.552 0.010729 *
ns(NUMERIC_AGE, df = 3)3 0.028585 0.079234 0.361 0.718276
PM_VISIT_LAST_2_YRS 0.197504 0.026124 7.560 4.21e-14 ***
log10plus1(VISIT_COUNT) 0.461070 0.026760 17.230 < 2e-16 ***
AF_25K_GIFT 0.685534 0.039273 17.456 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.436187 0.006143 71.011 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 0.424618 0.038353 11.071 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 -0.544680 0.023104 -23.575 < 2e-16 ***
MG_250K_PLUS 1.548409 0.065901 23.496 < 2e-16 ***
PRESIDENT_VISIT 0.195089 0.055185 3.535 0.000408 ***
TRUSTEE_OR_ADVISORY_BOARD 0.115450 0.029929 3.857 0.000115 ***
Alumnus -0.656304 0.021685 -30.265 < 2e-16 ***
EVER_PARENT -0.076955 0.019689 -3.908 9.32e-05 ***
SEASON_TICKET_YEARS -0.020693 0.004062 -5.094 3.53e-07 ***
CHICAGO_HOME 0.053747 0.015229 3.529 0.000418 ***
AFFINITY_SCORE 0.078956 0.006423 12.293 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier 0.083436 0.020040 4.163 3.15e-05 ***
MG_PR_MODEL_DESCMiddle Tier 0.086792 0.025718 3.375 0.000740 ***
MG_PR_MODEL_DESCTop Tier 0.371640 0.025589 14.524 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8695 on 18757 degrees of freedom
Multiple R-squared: 0.593, Adjusted R-squared: 0.5925
F-statistic: 1366 on 20 and 18757 DF, p-value: < 2.2e-16
[[5]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ns(NUMERIC_AGE,
df = 3) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + EVER_PARENT + SEASON_TICKET_YEARS + CHICAGO_HOME +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-2.8848 -0.6575 -0.0199 0.5743 4.2519
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.0886942 0.0661946 16.447 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)1 -0.4696901 0.0411656 -11.410 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)2 0.3604577 0.1407602 2.561 0.010451 *
ns(NUMERIC_AGE, df = 3)3 0.0005923 0.0801109 0.007 0.994100
PM_VISIT_LAST_2_YRS 0.2087323 0.0262004 7.967 1.72e-15 ***
log10plus1(VISIT_COUNT) 0.4605634 0.0269397 17.096 < 2e-16 ***
AF_25K_GIFT 0.6901754 0.0394121 17.512 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.4417796 0.0061375 71.981 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 0.4072144 0.0384912 10.579 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 -0.5532590 0.0232276 -23.819 < 2e-16 ***
MG_250K_PLUS 1.5738834 0.0659345 23.870 < 2e-16 ***
PRESIDENT_VISIT 0.1605536 0.0547808 2.931 0.003385 **
TRUSTEE_OR_ADVISORY_BOARD 0.1052592 0.0297605 3.537 0.000406 ***
Alumnus -0.6531787 0.0217434 -30.040 < 2e-16 ***
EVER_PARENT -0.0736060 0.0198622 -3.706 0.000211 ***
SEASON_TICKET_YEARS -0.0207069 0.0041205 -5.025 5.07e-07 ***
CHICAGO_HOME 0.0563207 0.0152364 3.696 0.000219 ***
AFFINITY_SCORE 0.0816554 0.0064175 12.724 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier 0.1074746 0.0200802 5.352 8.79e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.0985707 0.0256736 3.839 0.000124 ***
MG_PR_MODEL_DESCTop Tier 0.3775798 0.0255767 14.763 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8713 on 18757 degrees of freedom
Multiple R-squared: 0.5937, Adjusted R-squared: 0.5933
F-statistic: 1370 on 20 and 18757 DF, p-value: < 2.2e-16
[[6]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ns(NUMERIC_AGE,
df = 3) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + EVER_PARENT + SEASON_TICKET_YEARS + CHICAGO_HOME +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-2.8802 -0.6590 -0.0181 0.5746 4.2349
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.096866 0.065630 16.713 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)1 -0.464014 0.041073 -11.297 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)2 0.390313 0.139833 2.791 0.005255 **
ns(NUMERIC_AGE, df = 3)3 0.025561 0.079495 0.322 0.747807
PM_VISIT_LAST_2_YRS 0.197285 0.026368 7.482 7.65e-14 ***
log10plus1(VISIT_COUNT) 0.466281 0.026839 17.373 < 2e-16 ***
AF_25K_GIFT 0.668930 0.040000 16.723 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.442310 0.006157 71.836 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 0.419811 0.038558 10.888 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 -0.547008 0.023186 -23.592 < 2e-16 ***
MG_250K_PLUS 1.610948 0.066919 24.073 < 2e-16 ***
PRESIDENT_VISIT 0.194161 0.055657 3.489 0.000487 ***
TRUSTEE_OR_ADVISORY_BOARD 0.104165 0.029777 3.498 0.000470 ***
Alumnus -0.658298 0.021724 -30.302 < 2e-16 ***
EVER_PARENT -0.077621 0.019818 -3.917 9.01e-05 ***
SEASON_TICKET_YEARS -0.020973 0.004077 -5.144 2.72e-07 ***
CHICAGO_HOME 0.057194 0.015227 3.756 0.000173 ***
AFFINITY_SCORE 0.079260 0.006430 12.326 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier 0.092727 0.020085 4.617 3.92e-06 ***
MG_PR_MODEL_DESCMiddle Tier 0.091485 0.025729 3.556 0.000378 ***
MG_PR_MODEL_DESCTop Tier 0.363516 0.025582 14.210 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8706 on 18757 degrees of freedom
Multiple R-squared: 0.5939, Adjusted R-squared: 0.5934
F-statistic: 1371 on 20 and 18757 DF, p-value: < 2.2e-16
[[7]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ns(NUMERIC_AGE,
df = 3) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + EVER_PARENT + SEASON_TICKET_YEARS + CHICAGO_HOME +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-2.8671 -0.6593 -0.0219 0.5725 4.2503
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.104386 0.066531 16.600 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)1 -0.467928 0.041462 -11.286 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)2 0.346309 0.141336 2.450 0.014284 *
ns(NUMERIC_AGE, df = 3)3 -0.010039 0.079281 -0.127 0.899243
PM_VISIT_LAST_2_YRS 0.185758 0.026186 7.094 1.35e-12 ***
log10plus1(VISIT_COUNT) 0.470354 0.026838 17.525 < 2e-16 ***
AF_25K_GIFT 0.686516 0.039351 17.446 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.441887 0.006148 71.875 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 0.407849 0.038557 10.578 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 -0.542282 0.023231 -23.343 < 2e-16 ***
MG_250K_PLUS 1.642747 0.066931 24.544 < 2e-16 ***
PRESIDENT_VISIT 0.199382 0.055200 3.612 0.000305 ***
TRUSTEE_OR_ADVISORY_BOARD 0.109970 0.029659 3.708 0.000210 ***
Alumnus -0.652210 0.021829 -29.878 < 2e-16 ***
EVER_PARENT -0.079996 0.019914 -4.017 5.92e-05 ***
SEASON_TICKET_YEARS -0.020564 0.004155 -4.949 7.53e-07 ***
CHICAGO_HOME 0.058069 0.015237 3.811 0.000139 ***
AFFINITY_SCORE 0.078471 0.006426 12.212 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier 0.102345 0.020093 5.094 3.55e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.103764 0.025750 4.030 5.61e-05 ***
MG_PR_MODEL_DESCTop Tier 0.371402 0.025623 14.495 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8714 on 18757 degrees of freedom
Multiple R-squared: 0.5932, Adjusted R-squared: 0.5928
F-statistic: 1368 on 20 and 18757 DF, p-value: < 2.2e-16
[[8]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ns(NUMERIC_AGE,
df = 3) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + EVER_PARENT + SEASON_TICKET_YEARS + CHICAGO_HOME +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-2.8464 -0.6604 -0.0189 0.5745 4.2392
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.078159 0.066803 16.139 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)1 -0.423016 0.041456 -10.204 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)2 0.428920 0.141687 3.027 0.002471 **
ns(NUMERIC_AGE, df = 3)3 0.039286 0.078238 0.502 0.615577
PM_VISIT_LAST_2_YRS 0.205566 0.026278 7.823 5.44e-15 ***
log10plus1(VISIT_COUNT) 0.470443 0.026910 17.482 < 2e-16 ***
AF_25K_GIFT 0.653764 0.039774 16.437 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.437396 0.006167 70.929 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 0.414338 0.038399 10.790 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 -0.546444 0.023254 -23.499 < 2e-16 ***
MG_250K_PLUS 1.581509 0.066465 23.795 < 2e-16 ***
PRESIDENT_VISIT 0.180100 0.055670 3.235 0.001218 **
TRUSTEE_OR_ADVISORY_BOARD 0.100407 0.030129 3.333 0.000862 ***
Alumnus -0.645503 0.021801 -29.609 < 2e-16 ***
EVER_PARENT -0.084911 0.019853 -4.277 1.90e-05 ***
SEASON_TICKET_YEARS -0.017951 0.004092 -4.386 1.16e-05 ***
CHICAGO_HOME 0.052890 0.015269 3.464 0.000534 ***
AFFINITY_SCORE 0.080414 0.006428 12.511 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier 0.089575 0.020028 4.472 7.78e-06 ***
MG_PR_MODEL_DESCMiddle Tier 0.093402 0.025690 3.636 0.000278 ***
MG_PR_MODEL_DESCTop Tier 0.368437 0.025569 14.409 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8709 on 18757 degrees of freedom
Multiple R-squared: 0.5928, Adjusted R-squared: 0.5924
F-statistic: 1365 on 20 and 18757 DF, p-value: < 2.2e-16
[[9]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ns(NUMERIC_AGE,
df = 3) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + EVER_PARENT + SEASON_TICKET_YEARS + CHICAGO_HOME +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-2.8859 -0.6590 -0.0183 0.5733 4.2352
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.074699 0.066757 16.099 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)1 -0.421436 0.041473 -10.162 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)2 0.427513 0.141851 3.014 0.002583 **
ns(NUMERIC_AGE, df = 3)3 -0.001252 0.079080 -0.016 0.987367
PM_VISIT_LAST_2_YRS 0.190448 0.026024 7.318 2.61e-13 ***
log10plus1(VISIT_COUNT) 0.457093 0.026824 17.040 < 2e-16 ***
AF_25K_GIFT 0.664158 0.039490 16.818 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.441899 0.006158 71.756 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 0.409312 0.038655 10.589 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 -0.537054 0.023202 -23.147 < 2e-16 ***
MG_250K_PLUS 1.568594 0.066087 23.735 < 2e-16 ***
PRESIDENT_VISIT 0.157604 0.054862 2.873 0.004074 **
TRUSTEE_OR_ADVISORY_BOARD 0.100599 0.029793 3.377 0.000735 ***
Alumnus -0.651380 0.021727 -29.981 < 2e-16 ***
EVER_PARENT -0.077095 0.019867 -3.881 0.000105 ***
SEASON_TICKET_YEARS -0.021987 0.004084 -5.384 7.37e-08 ***
CHICAGO_HOME 0.051820 0.015244 3.399 0.000677 ***
AFFINITY_SCORE 0.078045 0.006417 12.163 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier 0.084092 0.020122 4.179 2.94e-05 ***
MG_PR_MODEL_DESCMiddle Tier 0.083721 0.025840 3.240 0.001198 **
MG_PR_MODEL_DESCTop Tier 0.379042 0.025543 14.840 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.871 on 18757 degrees of freedom
Multiple R-squared: 0.5945, Adjusted R-squared: 0.5941
F-statistic: 1375 on 20 and 18757 DF, p-value: < 2.2e-16
[[10]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ns(NUMERIC_AGE,
df = 3) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + EVER_PARENT + SEASON_TICKET_YEARS + CHICAGO_HOME +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-2.8682 -0.6606 -0.0198 0.5756 3.9778
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.069894 0.066558 16.075 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)1 -0.435309 0.041423 -10.509 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)2 0.425129 0.140997 3.015 0.002572 **
ns(NUMERIC_AGE, df = 3)3 0.033406 0.078366 0.426 0.669909
PM_VISIT_LAST_2_YRS 0.196509 0.026085 7.533 5.17e-14 ***
log10plus1(VISIT_COUNT) 0.457990 0.026868 17.046 < 2e-16 ***
AF_25K_GIFT 0.684323 0.039335 17.397 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.436112 0.006157 70.827 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 0.429306 0.038571 11.130 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 -0.526257 0.023222 -22.662 < 2e-16 ***
MG_250K_PLUS 1.588142 0.066259 23.969 < 2e-16 ***
PRESIDENT_VISIT 0.154460 0.055471 2.785 0.005366 **
TRUSTEE_OR_ADVISORY_BOARD 0.110727 0.029848 3.710 0.000208 ***
Alumnus -0.639238 0.021786 -29.341 < 2e-16 ***
EVER_PARENT -0.074822 0.019873 -3.765 0.000167 ***
SEASON_TICKET_YEARS -0.020460 0.004127 -4.957 7.22e-07 ***
CHICAGO_HOME 0.056929 0.015290 3.723 0.000197 ***
AFFINITY_SCORE 0.078509 0.006421 12.228 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier 0.099088 0.020160 4.915 8.95e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.091319 0.025805 3.539 0.000403 ***
MG_PR_MODEL_DESCTop Tier 0.375015 0.025618 14.639 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8722 on 18753 degrees of freedom
Multiple R-squared: 0.5917, Adjusted R-squared: 0.5912
F-statistic: 1359 on 20 and 18753 DF, p-value: < 2.2e-16Transaction all predictors 2
lapply(tlmaps2, function(x) summary(x))[[1]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ns(NUMERIC_AGE,
df = 3) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + EVER_PARENT + SEASON_TICKET_YEARS + CHICAGO_HOME +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-2.8750 -0.6584 -0.0198 0.5747 4.2474
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.106745 0.066358 16.678 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)1 -0.453626 0.041248 -10.998 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)2 0.358644 0.140827 2.547 0.010883 *
ns(NUMERIC_AGE, df = 3)3 0.026666 0.079578 0.335 0.737563
PM_VISIT_LAST_2_YRS 0.192421 0.026059 7.384 1.60e-13 ***
log10plus1(VISIT_COUNT) 0.474494 0.026865 17.662 < 2e-16 ***
AF_25K_GIFT 0.684029 0.039212 17.444 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.438473 0.006157 71.214 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 0.422669 0.038528 10.970 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 -0.541653 0.023189 -23.358 < 2e-16 ***
MG_250K_PLUS 1.602063 0.066422 24.119 < 2e-16 ***
PRESIDENT_VISIT 0.157782 0.055195 2.859 0.004259 **
TRUSTEE_OR_ADVISORY_BOARD 0.102029 0.029641 3.442 0.000578 ***
Alumnus -0.644498 0.021767 -29.609 < 2e-16 ***
EVER_PARENT -0.076859 0.019829 -3.876 0.000106 ***
SEASON_TICKET_YEARS -0.019776 0.004100 -4.823 1.42e-06 ***
CHICAGO_HOME 0.051622 0.015248 3.386 0.000712 ***
AFFINITY_SCORE 0.077855 0.006422 12.124 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier 0.098617 0.020053 4.918 8.83e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.097681 0.025684 3.803 0.000143 ***
MG_PR_MODEL_DESCTop Tier 0.370779 0.025548 14.513 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8707 on 18757 degrees of freedom
Multiple R-squared: 0.5928, Adjusted R-squared: 0.5924
F-statistic: 1365 on 20 and 18757 DF, p-value: < 2.2e-16
[[2]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ns(NUMERIC_AGE,
df = 3) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + EVER_PARENT + SEASON_TICKET_YEARS + CHICAGO_HOME +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-2.8821 -0.6584 -0.0213 0.5727 4.2596
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.088929 0.066440 16.390 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)1 -0.441453 0.040704 -10.845 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)2 0.375457 0.139812 2.685 0.007250 **
ns(NUMERIC_AGE, df = 3)3 -0.004023 0.073173 -0.055 0.956151
PM_VISIT_LAST_2_YRS 0.191826 0.026036 7.368 1.81e-13 ***
log10plus1(VISIT_COUNT) 0.452768 0.026840 16.869 < 2e-16 ***
AF_25K_GIFT 0.645790 0.039735 16.252 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.439840 0.006126 71.797 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 0.407928 0.038410 10.620 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 -0.558673 0.023165 -24.117 < 2e-16 ***
MG_250K_PLUS 1.617986 0.064628 25.035 < 2e-16 ***
PRESIDENT_VISIT 0.164765 0.054930 3.000 0.002708 **
TRUSTEE_OR_ADVISORY_BOARD 0.099861 0.029783 3.353 0.000801 ***
Alumnus -0.648234 0.021709 -29.861 < 2e-16 ***
EVER_PARENT -0.076365 0.019731 -3.870 0.000109 ***
SEASON_TICKET_YEARS -0.018710 0.004126 -4.535 5.79e-06 ***
CHICAGO_HOME 0.050880 0.015188 3.350 0.000810 ***
AFFINITY_SCORE 0.080413 0.006400 12.565 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier 0.102848 0.019979 5.148 2.66e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.119968 0.025612 4.684 2.83e-06 ***
MG_PR_MODEL_DESCTop Tier 0.402678 0.025612 15.722 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8683 on 18757 degrees of freedom
Multiple R-squared: 0.5968, Adjusted R-squared: 0.5963
F-statistic: 1388 on 20 and 18757 DF, p-value: < 2.2e-16
[[3]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ns(NUMERIC_AGE,
df = 3) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + EVER_PARENT + SEASON_TICKET_YEARS + CHICAGO_HOME +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-2.8771 -0.6623 -0.0163 0.5748 4.2590
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.082645 0.067339 16.078 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)1 -0.449336 0.041525 -10.821 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)2 0.377538 0.142841 2.643 0.008223 **
ns(NUMERIC_AGE, df = 3)3 0.015328 0.078600 0.195 0.845382
PM_VISIT_LAST_2_YRS 0.197041 0.026073 7.557 4.30e-14 ***
log10plus1(VISIT_COUNT) 0.464230 0.026836 17.299 < 2e-16 ***
AF_25K_GIFT 0.670878 0.039352 17.048 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.441327 0.006155 71.707 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 0.399131 0.038524 10.361 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 -0.555852 0.023280 -23.877 < 2e-16 ***
MG_250K_PLUS 1.611500 0.065798 24.492 < 2e-16 ***
PRESIDENT_VISIT 0.141530 0.054885 2.579 0.009926 **
TRUSTEE_OR_ADVISORY_BOARD 0.106714 0.029899 3.569 0.000359 ***
Alumnus -0.642949 0.021747 -29.565 < 2e-16 ***
EVER_PARENT -0.065420 0.019842 -3.297 0.000979 ***
SEASON_TICKET_YEARS -0.019819 0.004058 -4.884 1.05e-06 ***
CHICAGO_HOME 0.053847 0.015299 3.520 0.000433 ***
AFFINITY_SCORE 0.081961 0.006425 12.757 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier 0.098145 0.020068 4.891 1.01e-06 ***
MG_PR_MODEL_DESCMiddle Tier 0.097822 0.025746 3.800 0.000145 ***
MG_PR_MODEL_DESCTop Tier 0.378193 0.025645 14.747 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.872 on 18757 degrees of freedom
Multiple R-squared: 0.5955, Adjusted R-squared: 0.5951
F-statistic: 1381 on 20 and 18757 DF, p-value: < 2.2e-16
[[4]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ns(NUMERIC_AGE,
df = 3) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + EVER_PARENT + SEASON_TICKET_YEARS + CHICAGO_HOME +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-2.8916 -0.6551 -0.0172 0.5712 4.2223
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.131272 0.065896 17.168 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)1 -0.460029 0.041068 -11.202 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)2 0.357113 0.139953 2.552 0.010729 *
ns(NUMERIC_AGE, df = 3)3 0.028585 0.079234 0.361 0.718276
PM_VISIT_LAST_2_YRS 0.197504 0.026124 7.560 4.21e-14 ***
log10plus1(VISIT_COUNT) 0.461070 0.026760 17.230 < 2e-16 ***
AF_25K_GIFT 0.685534 0.039273 17.456 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.436187 0.006143 71.011 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 0.424618 0.038353 11.071 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 -0.544680 0.023104 -23.575 < 2e-16 ***
MG_250K_PLUS 1.548409 0.065901 23.496 < 2e-16 ***
PRESIDENT_VISIT 0.195089 0.055185 3.535 0.000408 ***
TRUSTEE_OR_ADVISORY_BOARD 0.115450 0.029929 3.857 0.000115 ***
Alumnus -0.656304 0.021685 -30.265 < 2e-16 ***
EVER_PARENT -0.076955 0.019689 -3.908 9.32e-05 ***
SEASON_TICKET_YEARS -0.020693 0.004062 -5.094 3.53e-07 ***
CHICAGO_HOME 0.053747 0.015229 3.529 0.000418 ***
AFFINITY_SCORE 0.078956 0.006423 12.293 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier 0.083436 0.020040 4.163 3.15e-05 ***
MG_PR_MODEL_DESCMiddle Tier 0.086792 0.025718 3.375 0.000740 ***
MG_PR_MODEL_DESCTop Tier 0.371640 0.025589 14.524 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8695 on 18757 degrees of freedom
Multiple R-squared: 0.593, Adjusted R-squared: 0.5925
F-statistic: 1366 on 20 and 18757 DF, p-value: < 2.2e-16
[[5]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ns(NUMERIC_AGE,
df = 3) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + EVER_PARENT + SEASON_TICKET_YEARS + CHICAGO_HOME +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-2.8848 -0.6575 -0.0199 0.5743 4.2519
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.0886942 0.0661946 16.447 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)1 -0.4696901 0.0411656 -11.410 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)2 0.3604577 0.1407602 2.561 0.010451 *
ns(NUMERIC_AGE, df = 3)3 0.0005923 0.0801109 0.007 0.994100
PM_VISIT_LAST_2_YRS 0.2087323 0.0262004 7.967 1.72e-15 ***
log10plus1(VISIT_COUNT) 0.4605634 0.0269397 17.096 < 2e-16 ***
AF_25K_GIFT 0.6901754 0.0394121 17.512 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.4417796 0.0061375 71.981 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 0.4072144 0.0384912 10.579 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 -0.5532590 0.0232276 -23.819 < 2e-16 ***
MG_250K_PLUS 1.5738834 0.0659345 23.870 < 2e-16 ***
PRESIDENT_VISIT 0.1605536 0.0547808 2.931 0.003385 **
TRUSTEE_OR_ADVISORY_BOARD 0.1052592 0.0297605 3.537 0.000406 ***
Alumnus -0.6531787 0.0217434 -30.040 < 2e-16 ***
EVER_PARENT -0.0736060 0.0198622 -3.706 0.000211 ***
SEASON_TICKET_YEARS -0.0207069 0.0041205 -5.025 5.07e-07 ***
CHICAGO_HOME 0.0563207 0.0152364 3.696 0.000219 ***
AFFINITY_SCORE 0.0816554 0.0064175 12.724 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier 0.1074746 0.0200802 5.352 8.79e-08 ***
MG_PR_MODEL_DESCMiddle Tier 0.0985707 0.0256736 3.839 0.000124 ***
MG_PR_MODEL_DESCTop Tier 0.3775798 0.0255767 14.763 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8713 on 18757 degrees of freedom
Multiple R-squared: 0.5937, Adjusted R-squared: 0.5933
F-statistic: 1370 on 20 and 18757 DF, p-value: < 2.2e-16
[[6]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ns(NUMERIC_AGE,
df = 3) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + EVER_PARENT + SEASON_TICKET_YEARS + CHICAGO_HOME +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-2.8802 -0.6590 -0.0181 0.5746 4.2349
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.096866 0.065630 16.713 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)1 -0.464014 0.041073 -11.297 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)2 0.390313 0.139833 2.791 0.005255 **
ns(NUMERIC_AGE, df = 3)3 0.025561 0.079495 0.322 0.747807
PM_VISIT_LAST_2_YRS 0.197285 0.026368 7.482 7.65e-14 ***
log10plus1(VISIT_COUNT) 0.466281 0.026839 17.373 < 2e-16 ***
AF_25K_GIFT 0.668930 0.040000 16.723 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.442310 0.006157 71.836 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 0.419811 0.038558 10.888 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 -0.547008 0.023186 -23.592 < 2e-16 ***
MG_250K_PLUS 1.610948 0.066919 24.073 < 2e-16 ***
PRESIDENT_VISIT 0.194161 0.055657 3.489 0.000487 ***
TRUSTEE_OR_ADVISORY_BOARD 0.104165 0.029777 3.498 0.000470 ***
Alumnus -0.658298 0.021724 -30.302 < 2e-16 ***
EVER_PARENT -0.077621 0.019818 -3.917 9.01e-05 ***
SEASON_TICKET_YEARS -0.020973 0.004077 -5.144 2.72e-07 ***
CHICAGO_HOME 0.057194 0.015227 3.756 0.000173 ***
AFFINITY_SCORE 0.079260 0.006430 12.326 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier 0.092727 0.020085 4.617 3.92e-06 ***
MG_PR_MODEL_DESCMiddle Tier 0.091485 0.025729 3.556 0.000378 ***
MG_PR_MODEL_DESCTop Tier 0.363516 0.025582 14.210 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8706 on 18757 degrees of freedom
Multiple R-squared: 0.5939, Adjusted R-squared: 0.5934
F-statistic: 1371 on 20 and 18757 DF, p-value: < 2.2e-16
[[7]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ns(NUMERIC_AGE,
df = 3) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + EVER_PARENT + SEASON_TICKET_YEARS + CHICAGO_HOME +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-2.8671 -0.6593 -0.0219 0.5725 4.2503
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.104386 0.066531 16.600 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)1 -0.467928 0.041462 -11.286 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)2 0.346309 0.141336 2.450 0.014284 *
ns(NUMERIC_AGE, df = 3)3 -0.010039 0.079281 -0.127 0.899243
PM_VISIT_LAST_2_YRS 0.185758 0.026186 7.094 1.35e-12 ***
log10plus1(VISIT_COUNT) 0.470354 0.026838 17.525 < 2e-16 ***
AF_25K_GIFT 0.686516 0.039351 17.446 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.441887 0.006148 71.875 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 0.407849 0.038557 10.578 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 -0.542282 0.023231 -23.343 < 2e-16 ***
MG_250K_PLUS 1.642747 0.066931 24.544 < 2e-16 ***
PRESIDENT_VISIT 0.199382 0.055200 3.612 0.000305 ***
TRUSTEE_OR_ADVISORY_BOARD 0.109970 0.029659 3.708 0.000210 ***
Alumnus -0.652210 0.021829 -29.878 < 2e-16 ***
EVER_PARENT -0.079996 0.019914 -4.017 5.92e-05 ***
SEASON_TICKET_YEARS -0.020564 0.004155 -4.949 7.53e-07 ***
CHICAGO_HOME 0.058069 0.015237 3.811 0.000139 ***
AFFINITY_SCORE 0.078471 0.006426 12.212 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier 0.102345 0.020093 5.094 3.55e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.103764 0.025750 4.030 5.61e-05 ***
MG_PR_MODEL_DESCTop Tier 0.371402 0.025623 14.495 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8714 on 18757 degrees of freedom
Multiple R-squared: 0.5932, Adjusted R-squared: 0.5928
F-statistic: 1368 on 20 and 18757 DF, p-value: < 2.2e-16
[[8]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ns(NUMERIC_AGE,
df = 3) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + EVER_PARENT + SEASON_TICKET_YEARS + CHICAGO_HOME +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-2.8464 -0.6604 -0.0189 0.5745 4.2392
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.078159 0.066803 16.139 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)1 -0.423016 0.041456 -10.204 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)2 0.428920 0.141687 3.027 0.002471 **
ns(NUMERIC_AGE, df = 3)3 0.039286 0.078238 0.502 0.615577
PM_VISIT_LAST_2_YRS 0.205566 0.026278 7.823 5.44e-15 ***
log10plus1(VISIT_COUNT) 0.470443 0.026910 17.482 < 2e-16 ***
AF_25K_GIFT 0.653764 0.039774 16.437 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.437396 0.006167 70.929 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 0.414338 0.038399 10.790 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 -0.546444 0.023254 -23.499 < 2e-16 ***
MG_250K_PLUS 1.581509 0.066465 23.795 < 2e-16 ***
PRESIDENT_VISIT 0.180100 0.055670 3.235 0.001218 **
TRUSTEE_OR_ADVISORY_BOARD 0.100407 0.030129 3.333 0.000862 ***
Alumnus -0.645503 0.021801 -29.609 < 2e-16 ***
EVER_PARENT -0.084911 0.019853 -4.277 1.90e-05 ***
SEASON_TICKET_YEARS -0.017951 0.004092 -4.386 1.16e-05 ***
CHICAGO_HOME 0.052890 0.015269 3.464 0.000534 ***
AFFINITY_SCORE 0.080414 0.006428 12.511 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier 0.089575 0.020028 4.472 7.78e-06 ***
MG_PR_MODEL_DESCMiddle Tier 0.093402 0.025690 3.636 0.000278 ***
MG_PR_MODEL_DESCTop Tier 0.368437 0.025569 14.409 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8709 on 18757 degrees of freedom
Multiple R-squared: 0.5928, Adjusted R-squared: 0.5924
F-statistic: 1365 on 20 and 18757 DF, p-value: < 2.2e-16
[[9]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ns(NUMERIC_AGE,
df = 3) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + EVER_PARENT + SEASON_TICKET_YEARS + CHICAGO_HOME +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-2.8859 -0.6590 -0.0183 0.5733 4.2352
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.074699 0.066757 16.099 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)1 -0.421436 0.041473 -10.162 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)2 0.427513 0.141851 3.014 0.002583 **
ns(NUMERIC_AGE, df = 3)3 -0.001252 0.079080 -0.016 0.987367
PM_VISIT_LAST_2_YRS 0.190448 0.026024 7.318 2.61e-13 ***
log10plus1(VISIT_COUNT) 0.457093 0.026824 17.040 < 2e-16 ***
AF_25K_GIFT 0.664158 0.039490 16.818 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.441899 0.006158 71.756 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 0.409312 0.038655 10.589 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 -0.537054 0.023202 -23.147 < 2e-16 ***
MG_250K_PLUS 1.568594 0.066087 23.735 < 2e-16 ***
PRESIDENT_VISIT 0.157604 0.054862 2.873 0.004074 **
TRUSTEE_OR_ADVISORY_BOARD 0.100599 0.029793 3.377 0.000735 ***
Alumnus -0.651380 0.021727 -29.981 < 2e-16 ***
EVER_PARENT -0.077095 0.019867 -3.881 0.000105 ***
SEASON_TICKET_YEARS -0.021987 0.004084 -5.384 7.37e-08 ***
CHICAGO_HOME 0.051820 0.015244 3.399 0.000677 ***
AFFINITY_SCORE 0.078045 0.006417 12.163 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier 0.084092 0.020122 4.179 2.94e-05 ***
MG_PR_MODEL_DESCMiddle Tier 0.083721 0.025840 3.240 0.001198 **
MG_PR_MODEL_DESCTop Tier 0.379042 0.025543 14.840 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.871 on 18757 degrees of freedom
Multiple R-squared: 0.5945, Adjusted R-squared: 0.5941
F-statistic: 1375 on 20 and 18757 DF, p-value: < 2.2e-16
[[10]]
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ns(NUMERIC_AGE,
df = 3) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + EVER_PARENT + SEASON_TICKET_YEARS + CHICAGO_HOME +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = .)
Residuals:
Min 1Q Median 3Q Max
-2.8682 -0.6606 -0.0198 0.5756 3.9778
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.069894 0.066558 16.075 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)1 -0.435309 0.041423 -10.509 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)2 0.425129 0.140997 3.015 0.002572 **
ns(NUMERIC_AGE, df = 3)3 0.033406 0.078366 0.426 0.669909
PM_VISIT_LAST_2_YRS 0.196509 0.026085 7.533 5.17e-14 ***
log10plus1(VISIT_COUNT) 0.457990 0.026868 17.046 < 2e-16 ***
AF_25K_GIFT 0.684323 0.039335 17.397 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.436112 0.006157 70.827 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 0.429306 0.038571 11.130 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 -0.526257 0.023222 -22.662 < 2e-16 ***
MG_250K_PLUS 1.588142 0.066259 23.969 < 2e-16 ***
PRESIDENT_VISIT 0.154460 0.055471 2.785 0.005366 **
TRUSTEE_OR_ADVISORY_BOARD 0.110727 0.029848 3.710 0.000208 ***
Alumnus -0.639238 0.021786 -29.341 < 2e-16 ***
EVER_PARENT -0.074822 0.019873 -3.765 0.000167 ***
SEASON_TICKET_YEARS -0.020460 0.004127 -4.957 7.22e-07 ***
CHICAGO_HOME 0.056929 0.015290 3.723 0.000197 ***
AFFINITY_SCORE 0.078509 0.006421 12.228 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier 0.099088 0.020160 4.915 8.95e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.091319 0.025805 3.539 0.000403 ***
MG_PR_MODEL_DESCTop Tier 0.375015 0.025618 14.639 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8722 on 18753 degrees of freedom
Multiple R-squared: 0.5917, Adjusted R-squared: 0.5912
F-statistic: 1359 on 20 and 18753 DF, p-value: < 2.2e-16Transaction final models
summary(tlm_final)
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ACTIVE_PROPOSALS +
AGE + PM_VISIT_LAST_2_YRS + VISITS_5PLUS + AF_25K_GIFT +
GAVE_IN_LAST_3_YRS + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DEEP_ENGAGEMENT + CHICAGO_HOME, data = mdat %>%
filter(rownum %in% unlist(xval_inds)))
Residuals:
Min 1Q Median 3Q Max
-3.3394 -0.5624 0.1473 0.7490 3.8297
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.76196 0.02216 79.521 < 2e-16 ***
ACTIVE_PROPOSALS 0.22505 0.03617 6.223 4.98e-10 ***
AGE 0.28640 0.01692 16.930 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.26428 0.03820 6.918 4.72e-12 ***
VISITS_5PLUS 0.72417 0.02747 26.363 < 2e-16 ***
AF_25K_GIFT 0.86000 0.04781 17.987 < 2e-16 ***
GAVE_IN_LAST_3_YRS 0.78335 0.01843 42.505 < 2e-16 ***
MG_250K_PLUS 1.76246 0.08019 21.978 < 2e-16 ***
PRESIDENT_VISIT 0.17698 0.06682 2.649 0.00808 **
TRUSTEE_OR_ADVISORY_BOARD 0.35546 0.03583 9.921 < 2e-16 ***
Alumnus 0.02143 0.02184 0.981 0.32654
DEEP_ENGAGEMENT 0.18696 0.01813 10.311 < 2e-16 ***
CHICAGO_HOME 0.13071 0.01762 7.419 1.22e-13 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 1.118 on 20851 degrees of freedom
Multiple R-squared: 0.3306, Adjusted R-squared: 0.3302
F-statistic: 858.1 on 12 and 20851 DF, p-value: < 2.2e-16summary(tlmap_final)
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ns(NUMERIC_AGE,
df = 3) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + EVER_PARENT + SEASON_TICKET_YEARS + CHICAGO_HOME +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = mdat %>% filter(rownum %in%
unlist(xval_inds)))
Residuals:
Min 1Q Median 3Q Max
-2.8816 -0.6585 -0.0194 0.5739 4.2452
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.086418 0.063348 17.150 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)1 -0.445402 0.039330 -11.325 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)2 0.399040 0.134481 2.967 0.003008 **
ns(NUMERIC_AGE, df = 3)3 0.017137 0.074762 0.229 0.818700
PM_VISIT_LAST_2_YRS 0.196261 0.024799 7.914 2.61e-15 ***
log10plus1(VISIT_COUNT) 0.463528 0.025472 18.198 < 2e-16 ***
AF_25K_GIFT 0.673509 0.037458 17.980 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.439683 0.005834 75.363 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 0.414306 0.036526 11.343 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 -0.545260 0.022014 -24.769 < 2e-16 ***
MG_250K_PLUS 1.594283 0.062723 25.418 < 2e-16 ***
PRESIDENT_VISIT 0.170416 0.052339 3.256 0.001132 **
TRUSTEE_OR_ADVISORY_BOARD 0.105537 0.028287 3.731 0.000191 ***
Alumnus -0.649499 0.020636 -31.474 < 2e-16 ***
EVER_PARENT -0.076141 0.018809 -4.048 5.18e-05 ***
SEASON_TICKET_YEARS -0.020165 0.003889 -5.185 2.18e-07 ***
CHICAGO_HOME 0.054331 0.014463 3.756 0.000173 ***
AFFINITY_SCORE 0.079521 0.006091 13.056 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier 0.095847 0.019039 5.034 4.84e-07 ***
MG_PR_MODEL_DESCMiddle Tier 0.096496 0.024403 3.954 7.70e-05 ***
MG_PR_MODEL_DESCTop Tier 0.375853 0.024275 15.483 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8708 on 20843 degrees of freedom
Multiple R-squared: 0.5938, Adjusted R-squared: 0.5934
F-statistic: 1523 on 20 and 20843 DF, p-value: < 2.2e-16Final models
lapply(final_models, function(x) summary(x))$`campaign.pg.score`
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ACTIVE_PROPOSALS +
AGE + PM_VISIT_LAST_2_YRS + VISITS_5PLUS + AF_25K_GIFT +
GAVE_IN_LAST_3_YRS + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DEEP_ENGAGEMENT + CHICAGO_HOME, data = mdat %>%
filter(rownum %nin% outliers))
Residuals:
Min 1Q Median 3Q Max
-4.8553 -1.2313 -0.1431 0.9948 5.6921
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.37197 0.02383 57.572 < 2e-16 ***
ACTIVE_PROPOSALS 0.27887 0.03914 7.125 1.07e-12 ***
AGE -0.22446 0.01815 -12.364 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.63584 0.04116 15.446 < 2e-16 ***
VISITS_5PLUS 0.72901 0.02954 24.675 < 2e-16 ***
AF_25K_GIFT 0.84388 0.05187 16.270 < 2e-16 ***
GAVE_IN_LAST_3_YRS 2.05061 0.01977 103.736 < 2e-16 ***
MG_250K_PLUS 1.04508 0.08572 12.191 < 2e-16 ***
PRESIDENT_VISIT 0.47405 0.07277 6.514 7.44e-11 ***
TRUSTEE_OR_ADVISORY_BOARD 0.27974 0.03841 7.282 3.38e-13 ***
Alumnus -0.14066 0.02353 -5.979 2.28e-09 ***
DEEP_ENGAGEMENT 0.31504 0.01946 16.190 < 2e-16 ***
CHICAGO_HOME 0.10443 0.01896 5.507 3.68e-08 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 1.344 on 26066 degrees of freedom
Multiple R-squared: 0.4644, Adjusted R-squared: 0.4641
F-statistic: 1883 on 12 and 26066 DF, p-value: < 2.2e-16
$largest.trans.pg.score
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ACTIVE_PROPOSALS +
AGE + PM_VISIT_LAST_2_YRS + VISITS_5PLUS + AF_25K_GIFT +
GAVE_IN_LAST_3_YRS + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + DEEP_ENGAGEMENT + CHICAGO_HOME, data = mdat %>%
filter(rownum %nin% outliers))
Residuals:
Min 1Q Median 3Q Max
-3.6439 -0.5621 0.1373 0.7348 4.2262
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.777602 0.019786 89.843 < 2e-16 ***
ACTIVE_PROPOSALS 0.214641 0.032496 6.605 4.05e-11 ***
AGE 0.296701 0.015073 19.685 < 2e-16 ***
PM_VISIT_LAST_2_YRS 0.286346 0.034178 8.378 < 2e-16 ***
VISITS_5PLUS 0.703268 0.024530 28.670 < 2e-16 ***
AF_25K_GIFT 0.870448 0.043064 20.213 < 2e-16 ***
GAVE_IN_LAST_3_YRS 0.781127 0.016413 47.593 < 2e-16 ***
MG_250K_PLUS 1.729127 0.071174 24.294 < 2e-16 ***
PRESIDENT_VISIT 0.141862 0.060419 2.348 0.0189 *
TRUSTEE_OR_ADVISORY_BOARD 0.359298 0.031894 11.265 < 2e-16 ***
Alumnus 0.007669 0.019534 0.393 0.6946
DEEP_ENGAGEMENT 0.187769 0.016156 11.622 < 2e-16 ***
CHICAGO_HOME 0.137207 0.015745 8.715 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 1.116 on 26066 degrees of freedom
Multiple R-squared: 0.3279, Adjusted R-squared: 0.3275
F-statistic: 1060 on 12 and 26066 DF, p-value: < 2.2e-16
$campaign.all.preds
Call:
lm(formula = log10plus1(CAMPAIGN_NEWGIFT_CMIT_CREDIT) ~ ns(NUMERIC_AGE,
df = 5) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + Alumnus + SEASON_TICKET_YEARS +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = mdat %>% filter(rownum %nin%
outliers))
Residuals:
Min 1Q Median 3Q Max
-4.3893 -0.6200 -0.1801 0.4947 6.2635
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.226749 0.091867 2.468 0.013585 *
ns(NUMERIC_AGE, df = 5)1 0.431494 0.084430 5.111 3.23e-07 ***
ns(NUMERIC_AGE, df = 5)2 0.311180 0.097686 3.186 0.001447 **
ns(NUMERIC_AGE, df = 5)3 -0.380150 0.062292 -6.103 1.06e-09 ***
ns(NUMERIC_AGE, df = 5)4 0.611299 0.206104 2.966 0.003020 **
ns(NUMERIC_AGE, df = 5)5 -0.362376 0.098792 -3.668 0.000245 ***
PM_VISIT_LAST_2_YRS 0.324460 0.024558 13.212 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.294325 0.024569 11.979 < 2e-16 ***
AF_25K_GIFT 0.609888 0.036933 16.513 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.261825 0.005764 45.425 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 2.917451 0.035980 81.085 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 0.574423 0.021716 26.452 < 2e-16 ***
MG_250K_PLUS 1.062765 0.060285 17.629 < 2e-16 ***
Alumnus -0.573615 0.018229 -31.467 < 2e-16 ***
SEASON_TICKET_YEARS -0.019746 0.003780 -5.223 1.77e-07 ***
AFFINITY_SCORE 0.155712 0.005786 26.913 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier -0.134420 0.018888 -7.117 1.13e-12 ***
MG_PR_MODEL_DESCMiddle Tier 0.181198 0.024139 7.507 6.26e-14 ***
MG_PR_MODEL_DESCTop Tier 0.562016 0.024047 23.372 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.9654 on 26060 degrees of freedom
Multiple R-squared: 0.7235, Adjusted R-squared: 0.7233
F-statistic: 3789 on 18 and 26060 DF, p-value: < 2.2e-16
$largest.trans.all.preds
Call:
lm(formula = log10plus1(LARGEST_GIFT_OR_PAYMENT) ~ ns(NUMERIC_AGE,
df = 3) + PM_VISIT_LAST_2_YRS + log10plus1(VISIT_COUNT) +
AF_25K_GIFT + sqrt(YEARS_OF_GIVING) + ns(YEARS_OF_GIVING_LAST_3,
df = 2) + MG_250K_PLUS + PRESIDENT_VISIT + TRUSTEE_OR_ADVISORY_BOARD +
Alumnus + EVER_PARENT + SEASON_TICKET_YEARS + CHICAGO_HOME +
AFFINITY_SCORE + MG_PR_MODEL_DESC, data = mdat %>% filter(rownum %nin%
outliers))
Residuals:
Min 1Q Median 3Q Max
-3.7286 -0.6589 -0.0182 0.5716 4.5864
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.056472 0.063350 16.677 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)1 -0.386585 0.036663 -10.544 < 2e-16 ***
ns(NUMERIC_AGE, df = 3)2 0.507589 0.134228 3.782 0.000156 ***
ns(NUMERIC_AGE, df = 3)3 0.052208 0.068724 0.760 0.447452
PM_VISIT_LAST_2_YRS 0.206842 0.022187 9.323 < 2e-16 ***
log10plus1(VISIT_COUNT) 0.465506 0.022757 20.456 < 2e-16 ***
AF_25K_GIFT 0.669998 0.033839 19.799 < 2e-16 ***
sqrt(YEARS_OF_GIVING) 0.435550 0.005218 83.473 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)1 0.414493 0.032630 12.703 < 2e-16 ***
ns(YEARS_OF_GIVING_LAST_3, df = 2)2 -0.538414 0.019672 -27.370 < 2e-16 ***
MG_250K_PLUS 1.571744 0.055798 28.169 < 2e-16 ***
PRESIDENT_VISIT 0.113327 0.047447 2.389 0.016923 *
TRUSTEE_OR_ADVISORY_BOARD 0.114033 0.025205 4.524 6.09e-06 ***
Alumnus -0.649690 0.018441 -35.232 < 2e-16 ***
EVER_PARENT -0.075523 0.016798 -4.496 6.96e-06 ***
SEASON_TICKET_YEARS -0.021058 0.003463 -6.080 1.22e-09 ***
CHICAGO_HOME 0.064236 0.012957 4.958 7.18e-07 ***
AFFINITY_SCORE 0.079900 0.005440 14.688 < 2e-16 ***
MG_PR_MODEL_DESCBottom Tier 0.080314 0.017056 4.709 2.51e-06 ***
MG_PR_MODEL_DESCMiddle Tier 0.075397 0.021789 3.460 0.000540 ***
MG_PR_MODEL_DESCTop Tier 0.355578 0.021708 16.380 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 0.8714 on 26058 degrees of freedom
Multiple R-squared: 0.5899, Adjusted R-squared: 0.5896
F-statistic: 1874 on 20 and 26058 DF, p-value: < 2.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